SRI LANKA JOURNAL OF ECONOMIC RESEARCHslfue.org/images/SLFUE_downloads/SLJER_Issues/2015June/... · 2016-09-20 · Sri Lanka Journal of Economic Research The International Journal

RESEARCH ARTICLES

The Impact of Currency Depreciation and Trade

Liberalization on Trade Balance of Sri Lanka

Nandasiri Keembiyahetti and Athula Naranpanawa

Education, Attitudes and Agricultural Biodiversity:

An application of Randomised Control Method

P.P.A. Wasantha Athukorala

Standardization and Characterization of Value Added

Watermelon Juice (Citrullus lanatus) Ready-to -Serve Beverage

K. Saranyah and T.Mahendran

The Solid Waste Management for Sustainable Development:

A Case Study of Hambantota Municipal Council Area in Sri Lanka

Asiri D. Vitharana

Unemployment, Official Economy and the Dimension of

the Shadow Economy: An Empirical Analysis for Sri Lanka

Using SEM Approach

D. I. J. Samaranayake and O.G. Dayarathna-Banda

PERSPECTIVES

Is Economics Independent of Morality?

Danny Atapattu

BOOK REVIEW

The Routledge Handbook of Environmental Economics in Asia,

Edited by Shunsuke Managi

J. G. Sri Ranjith

SRI LANKA

JOURNAL OF

ECONOMIC RESEARCH

Volume 3 Number 1 June 2015

Published on behalf of the SLFUE by

The Department of Economics, University of Ruhuna,

Matara, Sri Lanka

S L J E R

ISSN 2345-9913

Copyright © November 2015

SRI LANKA FORUM OF UNIVERSITY ECONOMISTS

National Library of Sri Lanka – Cataloguing-In-Publication Data

SRI LANKA JOURNAL OF ECONOMIC RESEARCH

The international journal of the Sri Lanka Forum of University

Economists

Volume 3 Number 1 (2015)

ISSN 2345 – 9913

Published by the Sri Lanka Forum of University Economists

The Editorial Office (2014) of the SLJER

The Department of Economics, University of Ruhuna, Matara, Sri Lanka.

Telephone: + 94 041 22 22681/Ext.3202

Web: www.slfue.org

Email: [email protected]

Editor (2014): A.J.M. Chandradasa

All rights reserved. No part of this publication may be reproduced, stored in

a retrieval system or transmitted by any means, electronic, mechanical,

photocopying, recording or otherwise, without written consent of the

publisher.

Printed by: Printec Establishment Private Limited

52/1, Ekwatta Road, Mirihana, Nugegoda, Sri Lanka

Telephone/Fax: +94 11 28 15 816

Web: www.printec.lk

Sri Lanka Journal of Economic Research

The International Journal of the Sri Lanka Forum of University Economists

Aims and Scope

Sri Lanka is currently facing many transitions: economic, epidemiological,

demographic, technological and social. The world economy too is evolving,

with technological progress, economic crises and social upheavals demanding

more and alternative economic analyses. Both these factors make it imperative

for economists in Sri Lanka and overseas, among the academic community as

well as practitioners, to focus on economic research and its dissemination. The

journal of the Sri Lanka Forum of University Economists seeks to fulfill this

mandate.

The Sri Lanka Journal of Economic Research (SLJER) creates a space where

research, particularly policy related research, can be disseminated and so

contributes to the economic thinking in the country in this period. The critical

evaluation of policy is essential if optimal use is to be made of the demographic

window of opportunity. Equity and social welfare, the cornerstone of

economic thinking in the country, and the challenges posed to such

fundamentals by economic liberalization, globalization and technological

progress make it vital to dwell on ideas and ideals, as well as to collate

systematic evidence to support rational policy making. The aim of this journal

then is to support such processes through dissemination and discussion.

The SLJER is a refereed tri-lingual journal. The journal will primarily provide

an opportunity for authors presenting papers at the annual sessions of the Sri

Lanka Forum of University Economists to disseminate their contributions.

Apart from the research articles the journal carries a special section titled

’Perspectives’ which articulates alternative thinking and approaches to

Economics. Book reviews are included as well.

All articles in this journal are subject to a rigorous blind peer-review process

initially, and are then reviewed by the editorial committee prior to final

acceptance for publication.

SPONSOR

This publication was fully sponsored by:

The National Council for Advanced

Studies in Humanities and Social

Sciences

64, Sukhasthan Gardens, Ward Place,

Colombo 7, Sri Lanka.

Telephone: +94 112 69 39 74

Fax: +94 112 69 39 75


Web: www.ncas.ac.lk

INQUIRIES

All inquiries to be directed to:

Secretariat of the SLJER (2012), Department of Economics, University of

Colombo, Kumaratunga Munidasa Mawatha, Colombo 3, Sri Lanka.

Telephone: +94 112 58 26 66 Fax: +94 112 50 27 22

Email: [email protected];

Secretariat (2013) of the Sri Lanka Forum of University Economists,

Department of Economics and Statistics, University of Peradeniya, Sri

Lanka;

Telephone: +94 812 39 26 22 Fax: 94 812 39 26 21


Secretariat (2014) of the Sri Lanka Forum of University Economists,

Department of Economics, University of Ruhuna, Sri Lanka;

Telephone: +94 (0) 41 22 22 681/Ext 32 02


Secretariat designate (2015) of the Sri Lanka Forum of University

Economists, Department of Business Economics, University of Sri

Jayawardenapura, Sri Lanka

Telephone: +94 (0) 28 02 005 Intercom: 8811


Board of Editors

A.J.M. Chandradasa (Chief Editor)

Department of Economics, University of Ruhuna, Sri Lanka

Danny Atapattu


Anushika Udayani kariyawasam

ELTUE, University of Ruhuna, Sri Lanka

Editorial Assistants

Udari Niranjala


A.J.M.U.S. Jayasundara,

Faculty of Applied Sciences, University of Sri Jayawardenapura, Sri Lanka

Editorial Advisory Board

W.D Lakshman

Emeritus Professor, University of Colombo, and Chairman, Institute of

Policy Studies of Sri Lanka, Colombo, Sri Lanka

D. Atapattu

Emeritus Professor, University of Ruhuna, Sri Lanka

Takao Fujimoto Professor of Economics,

Fukuoka University, Japan

Sarath Rajapathirana

Former Economic Adviser to the World Bank, and Former Member of the

Editorial Board of the World Bank Economic Review


ISSN 2345-9913

S L J E R Sri Lanka Journal of Economic Research


PANEL OF REVIEWERS

D. Atapattu

Senior Professor of Economics

University of Ruhuna, Sri Lanka

L.M. Abeywickrama

Professor of Agri-Economics


Athula Ranasinghe

Professor of Economics

University of Colombo, Sri Lanka

H.D. Karunarathne

Professor of Business Economics


U.K.J. Mudalige

Professor of Agri. Business

Wayamba University, Sri Lanka

A. Methsila Perera

Senior Lecturer (Accounting)

Wayamba University, Sri Lanka

O.G. Dayarathna Banda

Senior Lecturer (Economics)

University of Peradeniya, Sri Lanka

T. Lalithasiri Gunaruwan



K.A. Nisantha

Senior Lecturer (Social Statistics)


Vajira Bulugahapitiya

Senior Lecturer (Chemistry)





P.A.P. Samantha Kumara

Senior Lecturer (Marketing)


T.S.LW. Gunawardane

Senior Lecturer (Marketing)


J.M.A. Jayawickrama



P. Dunusinghe



H.M.W.A. Herath



K. Amirthalingam



Dr. B.M. Sumanarathne



RESEARCH ARTICLES

The Impact of Currency Depreciation and Trade 3

Liberalization on Trade Balance of Sri Lanka

Nandasiri Keembiyahetti and Athula Naranpanawa

Education, Attitudes and Agricultural Biodiversity: 29

An application of Randomised Control Method


Standardization and Characterization of Value Added 65

Watermelon Juice (Citrullus lanatus) Ready-to -Serve Beverage

K. Saranyah and T.Mahendran

The Solid Waste Management for Sustainable Development: 79

A Case Study of Hambantota Municipal Council Area in Sri Lanka

Asiri D. Vitharana

Unemployment, Official Economy and the Dimension of 113

the Shadow Economy: An Empirical Analysis for Sri Lanka

Using SEM Approach

D. I. J. Samaranayake and O.G. Dayarathna-Banda

PERSPECTIVES

Is Economics Independent of Morality? 157

Danny Atapattu

BOOK REVIEW

The Routledge Handbook of Environmental Economics in Asia, 169

Edited by Shunsuke Managi

J. G. Sri Ranjith

SRI LANKA

JOURNAL OF

ECONOMIC RESEARCH


Published on behalf of the SLFUE by

The Department of Economics, University of Ruhuna,

Matara, Sri Lanka

S L J E R

ISSN 2345-9913

RESEARCH ARTICLES

RESEARCH PAPAERS Sri Lanka Journal of

Economic Research

Volume 3 (1)

June 2015: 3-28

Sri Lanka Forum of

University Economists

SLJER

Nandasiri Keembiyahetti

Athula Naranpanawa

_______________________________________________

3

THE IMPACT OF CURRENCY

DEPRECIATION AND TRADE

LIBERRALZATION ON TRADE

BALANCE OF SRI LANKA

Abstract

This study examines the impact of trade liberalization and currency

depreciation on the trade balance of Sri Lanka. Using the bounds testing

ARDL approach for co-integration, which is more suitable for small finite

sample cases, we found that there was a long run co-integrating relationship

between the trade balance and its determinants; particularly the Trade

Openness and Real Exchange Rate. Our findings suggest that 1% increase in

trade openness leads to 0.48% deterioration, while 1 % depreciation of local

currency leads to 0.45% improvement in the Trade Balance Ratio of Sri

Lanka given all else remaining unchanged. These findings solve the

fundamental dilemma, why Sri Lanka’s Trade Balance continued to

deteriorate; despite of substantial currency devaluations/depreciations

allowed during past five decades. Our findings conclude that a more powerful

negative impact arising from trade openness fully offset the positive impact

arising from currency depreciation; thereby leading the Trade Balance into

deficit, eventually in the long run. Accordingly, we found trade liberalization

and devaluation are counter-cyclical as policy tools.

Keywords: – Trade Liberalization, Trade Balance, Bounds Testing, ARDL,

Sri Lanka JEL classification: C22; C51; F14

_____________________________________________________

Nandasiri Keembiyahetti Senior Lecturer, Department of Economics, Faculty of HSS, University of Ruhuna,

Matara, Sri Lanka. Email; [email protected]

Athula Naranpanawa Associate Professor, Department of Accounting, Finance and Economics, Griffith

Business School, Gold Coast Campus, Griffith University, Queensland 4222,

Australia. Tel: + 61 7 5552 8083 Fax: +61 7 5552 8068

E-mail address: [email protected]

mailto:[email protected]

SLJER Volume 3 Number 1, June 2015

4

INTRODUCTION

Trade Liberalization refers to the removal or reduction of artificial

barriers to trade goods and services among nations. This includes the

reduction or removal of tariff (import duties and surcharges) and non-

tariff barriers (licensing, quotas, rules of origin, exchange restrictions).

Trade liberalization on the one hand encourages countries to be

specialized in producing the goods and services, for which they have

comparative advantages. On the other hand the removal of barriers

would result affordably low prices for consumers. Moreover, trade

liberalization exposes local producers for greater competition

emerging from other nations. This would stimulate to increase

production efficiency, cost reduction or provide an incentive for an

industry to move resources into new ventures, not vulnerable to

competition. Trade Liberalization enables economies of scale and

greater specialization, particularly for small economies, having

geographical limitations in very own country.

Sometimes, Trade Liberalization might be threatening the balance of

an economy. Some industries may grow faster, some might be sluggish

or collapse; thereby causing structural unemployment. Trade

Liberalization can often be painful to less developed countries, having

a great deal of infant industries that cannot compete against foreign

competition. On the top of all, Trade Liberalization may create or

worsen the existing trade imbalances among nations.

The Impact of Currency Depreciation and Trade Liberalization

5

The effect of Trade Liberalization on the trade balance is theoretically

indeterminate. The Elasticity Approach suggests the effect will depend

on the price elasticity of demand for imports and exports, as well as the

extent to which the liberalization causes to change the relative price of

export and import. Though the relative prices could be enormously

changed due to tariff reduction under Trade Liberalization, it does not

assure improvement in trade balance unless Marshall-learner (ML)

condition; together with its assumptions is satisfied. Nevertheless,

‘Trade Openness’ is a broader concept beyond reduction of tariffs,

which involves removal of non-tariff barriers and exchange controls as

well. Hence, the Elasticity approach alone does not suffice to conclude

whether openness improves or worsens the trade balance. The

Absorption Approach to the balance of payments by Sydney Alexander

(Haberler, 1976) suggests the effect of liberalization will depend on

how real income is affected in relation to real absorption. A reduction

in import duties will shift expenditure to imported goods, thereby

raising foreign exchange outflows; but a reduction in export duties

does the opposite. Yet, the balance of payments will not improve if the

propensity to absorb is greater than unity.

Given this theoretical ambiguity, the impact of trade openness on the

trade balance becomes an empirical issue. ‘Trade Openness’ has

become an important policy variable for developing countries for the

last few decades, its impact on trade balance has recently received a

great deal of attention from researchers and policy makers alike; as


6

many developing countries continue to embark on Trade Liberalization

entering into bilateral, regional, and multilateral trade agreements. This

process was further encouraged by the World Bank and World Trade

Organization (WTO)

The objectives of this study are as follows: (a) To examine whether

long-run level relationship exists between trade balance of Sri Lanka

and its determinants using recent bound testing techniques (b) To

estimate the short-run and long-run elasticity of trade openness and to

exchange rate with respect to the trade balance using ARDL and EC

models (c) To evaluate trade openness and to exchange rate

devaluation as policy tools to rectify the persistent balance of trade

problem in Sri Lanka.

The remainder of this paper is organized as follows: beginning with a

brief literature review on ‘Trade Openness’ and balance of trade nexus,

it subsequently presents theoretical and econometric specification of

the model, Then, it discusses the empirical results, and finally,

concludes the findings and presents policy recommendations.

BRIEF LITERATURE REVIEW

Ostry and Rose (1992) using five different data sets (including one for

developing countries) found no statistically significant effect of tariff

changes on the real trade balance. Lutz and Singer (1994) addressed

the question of the effect of Trade Liberalization on a country’s term


7

of trade, based on a sample of 91 countries for 1968-88 including both

LDCs and industrialized countries. Their study conclude that

liberalization is likely to lead to the deterioration of the term of ‘trade,’

if Trade Liberalization was aimed at raising the size of the tradable

sector, either in absolute or relative terms. Santos-Paulino (2004) using

dynamic panel data and time series/cross-section analysis on the trade

balance and the current account balance of twenty two developing

countries from Latin America, Africa, East Asia and South Asia; where

significant trade reforms had been undertaken since the mid-1970s. It

has been found that liberalization has worsened the balance of trade

and the balance of payments; because imports have increased more

rapidly than exports. Using Bounds Testing Approach and Auto-

Regressive Distributed Lags (ARDL) model, Klasra (2011) tested the

existence of long run equilibrium relationship between the

determinants of growth during the period 1975–2004 for Pakistan and

Turkey. The results concluded that there is bi-directional causal

relationship between ‘Trade Openness’ and exports for Pakistan in the

short run. Moreover, they found that openness-growth nexus holds true

for Pakistan, and growth-driven exports hypothesis receives empirical

support for Turkey in the long run. Allaro (2012) examined the impact

of Trade Liberalization on the Ethiopia's trade balance, using the data

over the period 1974 to 2009; and found evidence to substantiate the

fact that Trade Liberalization led to deteriorate the trade balance due

to speedy increase in imports. Islam (2004) applied the Bounds Test to

determine the existence of a level relationship between government

sizes, openness, terms of trade volatility, and external risk using time


8

series data from Australia, Canada, England, Norway, Sweden, and the

US. Bounds Test results revealed that the existence of a long run

relationship in the US and Canada; but not in any of the other countries.

Among the studies pertaining to Sri Lanka, Chowdhury and Saleh

(2007) examined the long-run and short-run relationships between the

current account deficit, budget deficit, savings and investment gap and

‘Trade Openness’ in Sri Lanka using the Auto-Regressive Distributive

Lagged (ARDL) Approach. They found that ‘Trade Openness’ has a

positive effect on the current account deficit, but is statistically

insignificant. The result is not surprising, because the current account

balance includes private remittances from abroad; which is exogenous

to openness but helps in great deal to smooth the current account deficit

in Sri Lanka. Weliwita and Tsujii (2000) examined the responsiveness

of Sri Lanka’s trade deficit to currency devaluation during the post

liberalization period. The findings revealed that devaluation of Sri

Lankan rupee turned the trade deficit bad to worse, because trade

volumes were not responsive to the changes in real exchange rates. De

Silva and Zhu (2004) examined the effect of devaluation of rupee on

the trade balance and GDP using VAR and ECM, supplemented by

Variance Decomposition and Impulse Response Analysis. Their

findings based on aggregate annual data for 1977-1997, revealed that

the new exchange rate policy after 1977 has improved the trade

balance, but has failed to stimulate real output at least in the short run.


9

BRIEF OVERVIEW OF BALANCE OF TRADE BEHAVIOUR

IN SRI LANKA

A couple of fundamental macroeconomic problems that Sri Lanka has

been facing since the 1950s is the declining terms of trade and

widening trade gap. When the earnings from the traditional agricultural

exports of tea, rubber, and coconut had been weakening in the

international marketplace in the early 1960s, governments responded

to this situation by introducing tight regulations over foreign exchange.

Imports and exports were strongly regulated. This trend had been

accelerated from 1970 to1977, when a coalition headed by the Sri

Lanka Freedom Party imposed direct controls over international trade,

especially on imports. After its electoral victory in 1977, the United

National Party (UNP) government made enormous effort to liberalize

the economy and encouraged private enterprise, welcomed foreign

direct investment (FDI) and slackened import controls in a platform

called ‘Open Economic Policies.’ This is unambiguously accepted as

the historic turning point in trade linearization in Sri Lanka. Following

liberalization, Sri Lanka's economy became more diverse in the 1970s

and 1980s. In 1986 textiles and garments surpassed tea for the first time

as the country's ‘single largest export.’ Nevertheless, the performance

of the traditional agricultural exports remained essential to the

country's economic sustainability, while remittances from Sri Lankans

working overseas, foreign aids, and tourism became the other

important sources of foreign exchange.


10

From economic viewpoint, one of the most important benefits

associated with trade openness is the achievement of a faster economic

growth and development (Winters, 2004). The economic intuition

behind this statement is that less developed countries require acquiring

a huge amount of resources and exploring market opportunities for

domestically produced goods and services from advanced economies.

However, while Sri Lanka benefitted from Open Market Policies in

many different ways, its trade balance continued to deteriorate year by

year after introducing the ‘Open Economy’ in 1977

(15,000)

(10,000)

(5,000)

-

5,000

10,000

15,000

20,000

25,000

19

60

19

62

19

64

19

66

19

68

19

70

19

72

19

74

19

76

19

78

19

80

19

82

19

84

19

86

19

88

19

90

19

92

19

94

19

96

19

98

20

00

20

02

20

04

20

06

20

08

20

10

20

12

20

14

USD

Mn

Figure -1: Exports Imports and Trade Balance of Sri Lanka 1960-2014

(USD Million)

Export USD Mn

Import USD Mn

Trade Balance USD Mn

.

As shown in Figure-1, starting from US Dollar 41 Million surplus in

1977, the trade balance reported historic lowest downturn of US Dollar

9,710 Million by 2011, which is almost equal to the entire export

earnings of the country for that year. After three years later in 2014, it

has very marginally improved to US Dollar -8,285 Million which is

still as high as 74% of the total export earnings.


11

A careful study into the scatter plot depicted in Figure-2 would reveal

that Sri Lanka’s trade balance and exchange rate behaviour is totally

opposite to what is expected in ‘Economic Theory.’ Economic theory

postulates that trade balance should improve while exchange rate

continues to depreciate/devaluate given ML condition is satisfied.

Nevertheless, in Sri Lankan context the opposite seems true as

illustrated in Figure-2.

In this context, despite the studies so far done, we suspect that there

could be countercyclical effects between Trade Liberalization and

exchange devaluation on trade balance which offset the impact of each

other. As such, we attempt to answer the empirical question ‘why the

trade balance of Sri Lanka continued to deteriorate despite substantial

currency devaluations/depreciations during past five decades.’

-8,286 ,

131.05

-

20.00

40.00

60.00

80.00

100.00

120.00

140.00

-12,000 -10,000 -8,000 -6,000 -4,000 -2,000 - 2,000

Exch

an

ge

Ra

te L

KR

per

US

D

Trade Balance USD Mn

Figure-2, Scatter plot Nominal Exchange Rate vs.

Trade Balance of Sri Lanka 1960-2014


12

METHODOLOGY

Model

This study employs the standard “Two County” trade model as

stipulated by many studies in literature (see Rose and Yellen (1989);

Rose (1990); Bahmani-Oskooee (1991); Shirvani and Wilbratte

(1997); and Wilson (1999)) with a slight modification to incorporate

the ‘Trade Openness.’ The standard two country trade model assumes

that the demand for imports by domestic residents (Dm) depends on the

domestic income (y) and the relative price of imported goods to the

domestically produced goods (rpm) both expressed in home country

currency terms. Additionally, we assume that the trade balance is

affected by the degree of trade openness (top) of that economy. The

supper script f denotes the foreign counterpart of the analogous

variables. Thus, the initial equations can be given as:

Import demand

Dm = Dm(y, rpm, top) (1)

Dfm=Df

m(yf,rpfm,topf) (2)

rpm=e.pfx/p = (e.pf/p).(pf

x/pf) = q.rpfx (3)

q = e.pf/p (real exchange rate) (4)

(e- nominal exchange rate – the domestic currency price of foreign

exchange )


13

Analogously,

rpfm=(px/pf .e)= (p/pf.e).(px/p)=(1/q).rpx (5)

rpfm=rpx/q

export supply

Sx = Sx(rpx, top)

Sfx= Sf

x (rpfx,topf) (6)

Equilibrium condition

Dm=Sfx and Df

m=Sx (7)

TB = Real Trade balance of the domestic country

TB=rpx.Sx - rpm.Dm (8)

TB= rpx.Dfm - q.rpf

x.Dm

TB= rpx. Dfm(yf,rpf

m,topf) - q.rpfx. Dm(y, rpm, top)

TB= rpx. Dfm(yf, rpx/q,topf) - q.rpf

x. Dm(y, q.rpfx, top) (9)

Assume, no tariffs on exports, and no taxes or subsidies on

domestically trading goods; thus domestic price must be equal to

export price in any country; it means always rpx = rpfx =1 in equation

9; so equation becomes

TB= Dfm(yf, 1/q,topf) - q.Dm(y, q, top) (10)

Now TB is a function of

TB=TB(yf, y, q, top,topf)


14

Applying small country argument, we could ignore the impact of

foreign country’s ‘Trade Openness’ on the trade balance of the

domestic country, because a small country would demand (import) or

supply (export) a negligibly small volume which could be

exchangeable irrespective of the degree of trade openness of the rest of

the world. Hence, we can reasonably omit (topf ) from the above

function. Thus TB can be expressed as:

TB=TB(yf, y, q, top) (11)

This study applies the Bounds Testing Method, which i widely known

as the Autoregressive Distributed Lagged (ARDL) co-integration

procedure developed by Pesaran and Smith (2001) to analyse the long-

run relationships and dynamic interactions among the variables of

interest. ARDL procedure is more appropriate for this analysis due to

following reasons. Firstly, the Bounds Testing procedure does not

require the pre-testing of the variables in the model for unit roots. The

bounds testing procedure is applicable irrespective of whether the

regresses are integrated I(0), I(1) or mutually co-integrated, as long as

they are not co-integrated I(2). Secondly, the bounds testing procedure

is relatively more efficient in small or finite sample data, where the

number of observations by nature is small as in our case. Thirdly, both

short-run and long-run parameters can be estimated simultaneously.

Furthermore, as opposed to other multivariate co-integration

techniques such as Johansen and Juselius (1990), the Bounds Test

procedure is simple because the co-integration relationship can be

estimated using OLS; once the lag order of the model is defined by a

suitable information criterion.


15

Now we rewrite the functional relationship denoted in Eq(11) as

follows for notational convenience of the ARDL model.

𝑇𝐵 = 𝑓(𝐺𝐷𝑃, 𝐺𝐷𝑃𝑊, 𝑅𝐸𝑋, 𝑇𝑂𝑃, 𝑂𝐼𝐿, 𝑇) (12)

TB is the ratio of real exports to real imports both measured in USD

millions. Measuring TB as a ratio of exports to imports deserves at least

three advantages. It becomes unit less, non-negative allowing for log

transformation, and is normalized to one when trade is balanced.

GDP is Gross Domestic Product of home country and GDPW is the

sum of GDPs of the top 20 export destinations of Sri Lanka1.

REX is the real exchange defined in terms of domestic currency price

for one unit of US dollar. (Same as defined in Eq-4)

TOP is the sum of real exports and imports defined as a percentage of

GDP which is a proxy variable, representing the degree of Trade

Openness. Though it is subject to some limitations, this is the

conventional measure used to measure Trade Openness in many

empirical studies2. Many researchers use GDP in trade openness index

to normalize cross-sectional heterogeneity, depending on country size

which is not relevant to a single country case like ours. In our study,

taking total trade as a percentage of GDP at least produces a

1 U.S.A.,U.K., India, Italy, Belgium, Germany, Russia, U.A.E., Japan, Iran, France,

Netherlands, Canada, Turkey, Australia, China, Syria, Hong-Kong, Mexico,

Singapore, Switzerland. These countries account for 80.77% of Sri Lanka’s total

exports

2 See Islam(2004), Allaro (2012), Dritsaki (2013), Yanikkaya, H. (2003),

Narampanawa, A. (2011), Bowdler, C and Nunziata, L (2006),


16

comparable measure for Trade Openness over time. Nevertheless, this

measure as a proxy for Trade Openness is not free from errors. GDP

includes government expenditure (G) which is exogenously

determined, especially for country like Sri Lanka that had been

undergoing three decades of civil war. Thus, any abnormal increase in

G might erroneously means to suggest a reduction of Trade Openness,

as GDP appears in the denominator.

Also we included OIL dummy to capture world oil price hike in 1973

and a time (T) dummy for de-trending data.

The equation (12) was then transformed to an Auto-Regressive

Distributed Lag model (ARDL) to identify the existence of a long-run

co-integration relationship. TB, GDP, REX, TOP were considered

endogenous and GDPW and dummy variables were taken as

exogenous variables. The 𝑙𝑛 notation stands for natural logarithm

and 𝑒𝑡for the error term which is white noise.

𝑙𝑛𝑇𝐵𝑡 = 𝛽0𝑇𝐵 + 𝛽1𝑇𝐵𝑇𝑡 + 𝛽2𝑇𝐵𝑂𝐼𝐿𝑡 + 𝛽3𝑇𝐵𝑙𝑛𝐺𝐷𝑃𝑊𝑡 +

∑ 𝛼1𝑇𝐵𝑛𝑖=1 𝑙𝑛𝑇𝐵𝑡−𝑖 + ∑ 𝛼2𝑇𝐵

𝑛𝑖=0 𝑙𝑛𝐺𝐷𝑃𝑡−𝑖 + ∑ 𝛼3𝑇𝐵

𝑛𝑖=0 𝑙𝑛𝑅𝐸𝑋𝑡−𝑖 +

∑ 𝛼4𝑇𝐵𝑛𝑖=0 𝑙𝑛𝑇𝑂𝑃𝑡−𝑖 + 𝜀𝑇𝐵𝑡 (13)

𝑙𝑛𝐺𝐷𝑃𝑡 = 𝛽0𝐺𝐷𝑃 + 𝛽1𝐺𝐷𝑃𝑇𝑡 + 𝛽2𝐺𝐷𝑃𝑂𝐼𝐿𝑡 + 𝛽3𝐺𝐷𝑃𝑙𝑛𝐺𝐷𝑃𝑊𝑡 +

∑ 𝛼1𝐺𝐷𝑃𝑛𝑖=1 𝑙𝑛𝐺𝐷𝑃𝑡−𝑖 + ∑ 𝛼2𝐺𝐷𝑃

𝑛𝑖=0 𝑙𝑛𝑇𝐵𝑡−𝑖 +

∑ 𝛼3𝐺𝐷𝑃𝑛𝑖=0 𝑙𝑛𝑅𝐸𝑋𝑡−𝑖 + ∑ 𝛼4𝐺𝐷𝑃

𝑛𝑖=0 𝑙𝑛𝑇𝑂𝑃𝑡−𝑖 + 𝜀𝐺𝐷𝑃𝑡 (14)


17

𝑙𝑛𝑅𝐸𝑋𝑡 = 𝛽0𝑅𝐸𝑋 + 𝛽1𝑅𝐸𝑋𝑇𝑡 + 𝛽2𝑅𝐸𝑋𝑂𝐼𝐿𝑡 + 𝛽3𝑅𝐸𝑋𝑙𝑛𝐺𝐷𝑃𝑊𝑡 +

∑ 𝛼1𝑅𝐸𝑋𝑛𝑖=1 𝑙𝑛𝑅𝐸𝑋𝑡−𝑖 + ∑ 𝛼2𝑅𝐸𝑋

𝑛𝑖=0 𝑙𝑛𝐺𝐷𝑃𝑡−𝑖 +

∑ 𝛼3𝑅𝐸𝑋𝑛𝑖=0 𝑙𝑛𝑇𝐵𝑡−𝑖 + ∑ 𝛼4𝑅𝐸𝑋

𝑛𝑖=0 𝑙𝑛𝑇𝑂𝑃𝑡−𝑖 + 𝜀𝑅𝐸𝑋𝑡 (15)

𝑙𝑛𝑇𝑂𝑃𝑡 = 𝛽0𝑇𝑂𝑃 + 𝛽1𝑇𝑂𝑃𝑇𝑡 + 𝛽2𝑇𝑂𝑃𝑂𝐼𝐿𝑡 + 𝛽3𝑇𝑂𝑃𝑙𝑛𝐺𝐷𝑃𝑊𝑡 +

∑ 𝛼1𝑇𝑂𝑃𝑛𝑖=1 𝑙𝑛𝑇𝑂𝑃𝑡−𝑖 + ∑ 𝛼2𝑇𝑂𝑃

𝑛𝑖=0 𝑙𝑛𝐺𝐷𝑃𝑡−𝑖 +

∑ 𝛼3𝑇𝑂𝑃𝑛𝑖=0 𝑙𝑛𝑅𝐸𝑋𝑡−𝑖 + ∑ 𝛼4𝑇𝑂𝑃

𝑛𝑖=0 𝑙𝑛𝑇𝐵𝑡−𝑖 + 𝜀𝑇𝑂𝑃𝑡 (16)

Hypotheses

In equation 13, where real trade balance is the dependent variable

𝑓𝑇𝐵(𝑇𝐵|𝐺𝐷𝑃, 𝑅𝐸𝑋, 𝑇𝑂𝑃) the null hypothesis of no cointegration

amongst the variables is 𝐻0: ∝1𝑇𝐵= ∝2𝑇𝐵=∝3𝑇𝐵=∝4𝑇𝐵= 0 against the

alternative hypothesis 𝐻1: ∝1𝑇𝐵≠∝2𝑇𝐵≠∝3𝑇𝐵≠∝4𝑇𝐵≠0.

In equation 14, where GDP is the dependent variable,

𝑓𝐺𝐷𝑃(𝐺𝐷𝑃|𝑇𝐵, 𝑅𝐸𝑋, 𝑇𝑂𝑃) the null hypothesis of no cointegration is

𝐻0: ∝1𝐺𝐷𝑃= ∝2𝐺𝐷𝑃=∝3𝐺𝐷𝑃=∝4𝐺𝐷𝑃= 0 against the alternative

hypothesis 𝐻1: ∝1𝐺𝐷𝑃≠∝2𝐺𝐷𝑃≠∝3𝐺𝐷𝑃≠∝4𝐺𝐷𝑃≠0.

In equation 15, where REX is the dependent variable,

𝑓𝑅𝐸𝑋(𝑅𝐸𝑋|𝑇𝐵, 𝐺𝐷𝑃, 𝑇𝑂𝑃) the null hypothesis of no cointegration is

𝐻0: ∝1𝑅𝐸𝑋= ∝2𝑅𝐸𝑋=∝3𝑅𝐸𝑋=∝4𝑅𝐸𝑋= 0 against the alternative

hypothesis 𝐻1: ∝1𝑅𝐸𝑋≠∝2𝑅𝐸𝑋≠∝3𝑅𝐸𝑋≠∝4𝑅𝐸𝑋≠0.


18

Similarly in equation 16, where TOP is the dependent variable,

𝑓𝑇𝑂𝑃(𝑇𝑂𝑃|𝑇𝐵, 𝐺𝐷𝑃, 𝑅𝐸𝑋) the null hypothesis of no cointegration is

𝐻0: ∝1𝑇𝑂𝑃= ∝2𝑇𝑂𝑃=∝3𝑇𝑂𝑃=∝4𝑇𝑂𝑃= 0 against the alternative

hypothesis 𝐻1: ∝1𝑇𝑂𝑃≠∝2𝑇𝑂𝑃≠∝3𝑇𝑂𝑃≠∝4𝑇𝑂𝑃≠0.

Data

This study uses annual data1960-2014 from two sources. GDPW is

from author’s calculation with GDP data from International Financial

Statistics (IFS). All the other data are from the Annual Reports of the

Central Bank of Sri Lanka. GDP and export import data series were

rebased to 2005 constant price using relevant price indices.

EMPIRICAL RESULTS

The ARDL bounds test developed by Pesaran et al. (2001) can be used

to establish the short-run and long-run relationships, irrespective of

whether they are purely I(0), purely I(1), or mutually co-integrated.

Nevertheless, this procedure requires the variables under consideration

are not integrated at an order higher than one. In the presence of I(2)

variables, the model crashes because the critical values provided by

Pesaran et al. (2001) are no longer valid. Therefore, it is of crucial

importance to test the order of integration of all variables and to verify

none of them are of I(2). To test the null hypothesis of unit root against

the alternative of stationary, we performed Augmented Dickey–Fuller

(ADF) test for both “intercept only” and “intercept with liner trend”

methods but reported only the latter for brevity. The latter is more


19

reliable, as all the variables are trending over time and we have no valid

statistical reason to avoid intercept.

As per the ADF test results reported in Table-1, all the variables were

found I(1) in levels, but I(0) in first difference and more importantly,

no evidences were detected for I(2). Hence, the variables are qualified

to be used in bounds testing.

Table-1: Augmented Dickey-Fuller Unit Root Test (Intercept and Trend)

Variable Level First Differenced

t-statistic Probability t-statistic Probability

lnGDP -2.070 (5) 0.5478 -5.707(0)*** 0.0001

lnTB -1.831 (0) 0.6745 -8.581 (0)*** 0.0000

lnTOP -2.631 (0) 0.2689 -3.763 (10)** 0.0296

lnGDPW -2.396 (1) 0.3771 -5.733 (1)*** 0.0001

lnREX -0.826 (2) 0.9558 -5.265 (1)*** 0.0004

For ADF test within brackets are lag length selected by AIC

*** Significant at 1% level

** Significant at 5% level

In the bounds testing approach to ARDL, F-statistic is used to test the

existence of long-run relationship. The F-stat used for this procedure,

however, does not follow standard F-distribution. Thus, the Pesaran et

al. (1996) computed two sets of critical values for any given

conventional significance level. One set assumes that all variables are

integrated order I(0), and the other set assumes that they all are

integrated order I(1). If the calculated F-statistics exceeds, the upper

bound critical value, then the H0 of no co-integration will be rejected

favouring the alternative that co-integration among the variables. On

the other hand, if the calculated F-statistic is less than the lower bound


20

critical value, then H0 of no co-integration among the variables cannot

be rejected. If F-statistic falls within the bounds, then the test result is

inconclusive and existence of long-run relationship is indeterminate.

Table-2 shows the Result of Bound Testing for Co-integration in

ARDL Model. The results indicate that the null hypothesis of no co-

integration cannot be rejected for the models defined by Eq(14), Eq(15)

and Eq(16). It reveals that the calculated F-statistic exceeds the upper

bound critical value at 5% significant level only for model defined by

Eq(13), leading to the concussion that there exists only one long run

co-integrating relationship where TB appears being the dependent

variable.

Table-2 Result of Bound Testing for Cointegration in ARDL Model

using Akaike Information Criterion

ARDL Model ARDL

Lag

Length

F-

Statistics

Presence of

Cointegration

𝐹𝑇𝐵(𝑇𝐵|𝑇𝑂𝑃, 𝐺𝐷𝑃, 𝑅𝐸𝑋) 1,2,2,1 6.6842** YES

𝐹𝑇𝑂𝑃(𝑇𝑂𝑃|𝑇𝐵, 𝐺𝐷𝑃, 𝑅𝐸𝑋, ) 1,0,2,1 1.6558* NO

𝐹𝐺𝐷𝑃(𝐺𝐷𝑃|𝑇𝑂𝑃, 𝑇𝐵, 𝑅𝐸𝑋) 2,0,1,2 5.0474* NO

𝐹𝑅𝐸𝑋(𝑅𝐸𝑋|𝐺𝐷𝑃, 𝑇𝑂𝑃, 𝑇𝐵) 1,1,2,1 1.8186* NO

** Above the 95% Upper bound critical value

* Below the 95% Lower bound critical value

Note: The critical value bounds are computed by stochastic

simulations using 20000 replications. For this analysis 95%

critical bounds are 5.0376-6.2037

Having established the co-integrating relationship, the long run

coefficients for Eq(13) was estimated using ARDL (1,2,2,1) selected

based on Akaike Information Criterion. The estimated long run

coefficients are reported in Table-3


21

Table-3, The Estimated Long-run Coefficients

Regressor Coefficient Standard

Error

T-Ratio Prob

C -16.3545 22.6664 -0.7215 0.4760

T -0.0362 0.07406 -0.4899 0.6280

OIL 0.1966 0.17643 1.1141 0.2740

lnTOP -0.4803 ** 0.2281 -2.1055 0.0430

lnGDP 1.0655 1.4900 0.7151 0.4800

lnGDPW 0.0664 1.1923 0.0556 0.9560

lnREX 0.4596*** 0.1015 4.5240 0.0000

As shown in Table-3, all the variables take expected sign but only the

Trade Openness (TOT) and Real Exchange Rate (REX) are significant

in explaining long run variations in the Trade Balance (TB). The

estimated coefficient for REX shows highly significant long run impact

of Real Exchange Rate on the Trade Balance. The estimated coefficient

indicates that 1 % depreciation of local currency leads to 0.45%

improvement in the Trade Balance Ratio of Sri Lanka, given all else

being equal. On contrary the Trade Openness is negatively significant

at 5% level indicating that 1% increase in Trade Openness leads to

deteriorate the Trade Balance Ratio by 0.48%. The underlining

economic intuition is that the nature of Trade Openness in Sri Lanka

can be regarded as more supportive to imports than to exports. These

findings solve our basic puzzle that why Sri Lanka’s Trade Balance

continuous to deteriorate in long term despite of substantial currency

depreciation. It is proven in our findings that relatively a more

powerful negative impact arising from Trade Openness fully offset the

positive impact arising from currency depreciation ultimately leading

the Trade Balance into deficit in the long run.


22

Table-4, Error Correction Representation ARDL(1,2,2,1) selected

based on Akaike Information Criterion

Dependent variable is lnLTB

Regressor Coefficient Standard

Error

T-Ratio Prob

lnTOP -0.07473 0.17983 -0.4156 0.680

lnTOP1 0.44084 0.17988 2.4507 0.020

lnGDP -0.23013 1.01890 -0.2259 0.823

lnGDP1 -2.23960 1.00130 -2.2368 0.032

lnREX -0.07579 0.15949 -0.4752 0.638

T -0.02230 0.04366 -0.5108 0.613

OIL 0.12080 0.10365 1.1654 0.252

lnGDPW 0.04078 0.73425 0.0555 0.956

ECM(-1) -0.61456 0.14476 -4.2453 0.000

Notes, lnLTB = lnTB-lnTBt-1

lnTOP = lnTOP- TOPt-1

lnTOP1 = lnTOPt-1- lnTOPt-2

lnLGDP= lnGDP- lnGDPt-1

lnLGDP1= lnGDPt-1-lnGDPt-2

lnLREX = lnREX-lnREXt-1

T = T-Tt-1

OIL = OIL-OILt-1

lnLGDPW= lnGDPW-lnGDPWt-1

R-Squared 0.54956

R-Bar-Squared 0.37520

S.E. of Regression 0.091253

F-Stat. F(9,34) 4.2025[.001]

Residual Sum of Squares 0.25814

DW-statistic 2.3615

The results of the Error Correction Model (ECM) denotes the short-run

dynamics associated with the long run relationships are given in Table-

4. The short run dynamics tabulated in Table-4 suggest neither Real

Exchange Rate nor Trade Openness has significant influence on the

Trade Balance in the short run. However, the Error Correction Term

(ECM) estimated -0.61456 (0.000) is highly significant and negative

in sign indicating fairly a higher speed of adjustment towards

equilibrium after a short run shock. It is estimated approximately 61%

of disequilibria from the previous year would converge back to long

run equilibrium in the current year.


23

Figure-3 depicts the dynamic forecast for the long run path of Real

Trade Balance while Figure-4 and Figure-5 respectively depict the plot

of cumulative sum of recursive residuals (CUSUM) and the plot of

cumulative sum of squares recursive residuals (CUSUMQ).

Figure-3

Both CUSUM and CUSUMQ moving within critical bounds at 5%

significant level illustrate the model stability for the sample period.

-1.6

-1.4

-1.2

-1.0

-0.8

-0.6

-0.4

-0.2

1962 1974 1986 1998 2010

Dynamic forecasts for the level of LREALTB

LREALTB Forecast


24

Figure-4

Figure-5

-20

-10

0

10

20

1962 1973 1984 1995 2005

The straight lines represent critical bounds at 5% significance level

Plot of Cumulative Sum of Recursive Residuals

-0.5

0.0

0.5

1.0

1.5

1962 1973 1984 1995 2005

The straight lines represent critical bounds at 5% significance level

Plot of Cumulative Sum of Squares of Recursive Residuals


25

CONCLUSION

Employing the Bound Testing (ARDL) approach, this study examined

the short run and long run relationship between ‘Trade Openness’ and

the ‘Trade Balance’ of Sri Lanka. The Bounds test reveals that there is

a long run co-integrating relationship between the variables of interest,

predominantly the Trade Openness and Real Exchange Rate. Our

findings suggest that 1% increase in trade openness leads to 0.48%

deterioration while 1 % depreciation of local currency leads to 0.45%

improvement in the Trade Balance Ratio of Sri Lanka, given all else

remaining unchanged. These findings solve our basic dilemma, why

Sri Lanka’s Trade Balance continued to deteriorate, despite the

substantial currency depreciation allowed during last few decades. It is

proven that relatively a more powerful negative impact arising from

Trade Openness fully offset the positive impact arising from currency

depreciation; thereby leading the Trade Balance into deficit eventually

in the long run. Therefore, devaluating currency and liberalizing trade

are countercyclical policies offsetting the impact of each other. It leads

to the policy recommendation that government must allow further

depreciation of local currency if liberalizing trade any longer, in order

to prevent TB going bad to worse.

REFERENCES

Allaro, H. B. (2012). “The Impact of Trade Liberalization on the

Ethiopia's Trade Balance”, American Journal of Economics,

Vol 2(5): pp 75-81


26

Bahmani-Oskoee, M and Brooks T.J, (1991) “Cointegration approach

to Estimating Bilateral Trade elasticities between US and her

Trading partners” International Economic Journal, Vol.13, (4)

winter' pp 66-79

Bowdler, C and Nunziata, L (2006). “Trade Openness and Inflation

Episodes in the OECD”, Journal of Money, Credit and Banking,

Vol. 38 (2), pp. 553-563

Chowdhury, K. & Saleh, A.S. (2007). “Testing the Keynesian

Proposition of Twin Deficits in the Presence of Trade

Liberalization: Evidence from Sri Lanka” Working Paper 07-

09, Department of Economics, University of Wollongong,

2007. http://www.uow.edu.au/commerce/econ/wpapers.html

De Silva, D., Zhu, Z. (2004). “Sri Lanka’s Experiment with

Devaluation: VAR and ECM Analysis of the Exchange Rate

Effects on Trade Balance and GDP.” The International Trade

Journal 28(4): 269-301.

Dritsaki, M. & Dritski,C. (2013). “Bound Testing Approach for

Cointegration and Causality between Financial Development,

Trade Openness and Economic Growth in Bulgaria, the IUP

Journal of Applied Economics, Vol. 12 (1), pp. 50-67

Haberler, Gottfried (1976). "The Monetary Approach to the Balance of

Payments". Journal of Economic Literature 14 (4): pp. 1324–

1328

Islam, M. Q. (2004). “The long run relationship between openness and

government size: evidence from bounds test”, Applied

Economics, Vol. 36, pp. 995–1000

Johansen, S. and K. Juselius, (1990). “Maximum likelihood estimation

and inference on cointegration-with application to the demand

for money”. Oxford Bulletin of Economics and Statistics, Vol.

52 pp169-210


27

Klasra, M. A. (2011). ‘Foreign direct investment, trade openness and

economic growth in Pakistan and turkey: an investigation

using bounds test’ Springer Science Business Media B.V.

Qual Quant (published on line) 45: pp. 223–231

LUTZ, M. and Singer, H.W. (1994). “The Link Between Increased

Trade Openness and the Terms of Trade: An Empirical

Investigation” World Development, Vol. 22, No. 11, pp.

1697-1709

Narampanawa, A. (2011). “Does trade openness promote carbon

emissions? Empirical evidence from Sri Lanka”, The Empirical

Economic Letters, Vol. 10(10). pp. 973-986

Ostry, J. D. and Rose, A. K. (1992). ‘An Empirical Evaluation of the

Macroeconomic Effects of Tariffs’, Journal of International

Money and Finance, Vol. 11, pp. 63–79

Pesaran, M.H., Shin, Y. and Smith, R.J. (2001). “Bounds testing

approaches to the analysis of level relationships”. Journal of

Applied Econometrics, Vol.16. pp 289-326.

Rose, Andrew K, and Yellen, Janet.L. (1989), “Is There a J-Curve?”

Journal of Monetary Economics, vol. (24), pp 53-68.

Rose, Andrew K., (1990), “Exchange Rate and the Trade balance:

Some Evidence from Developing Countries”, Economic

Letters, pp 217-275

Santos-Paulino, A.U. (2004) ‘Trade Liberalization and the Balance of

Payments in Selected Developing Countries’, The Manchester

School Vol 72 No. , pp.100-118

Shirvani, H., & Wilbratte, B. (1997). The relationship between the real

exchange rate and the trade balance: An empirical

reassessment. International Economic Journal, Vol. 11(1),

pp.39-51.


28

Weliwita, A., Tsujii, H. (2000). “The Exchange Rate and Sri Lanka’s

Trade Deficit.” Journal of Economic Development 25: 131-

157.

Wilson, P. (1999). “Exchange Rates and the Trade Balance: the case

of Malaysia from 1970 to 1996.” Malaysian Journal of

Economic Studies 36(2): 1-15.

Winters, L. A. (2004). “Trade Liberalization and Economic

Performance: An Overview”. The Economic Journal, 114, F4-

F21.

Yanikkaya, H. (2003). “Trade Openness and Economic Growth: a

Cross Country Empirical Investigation”, Journal of

development Economics, Vol.72, pp.57-89


Economic Research

Volume 3 (1)

June 2015: 29-64

Sri Lanka Forum of


SLJER


_____________________________________________

29

EDUCATION, ATTITUDES AND

AGRICULTURAL BIODIVERSITY: AN

APPLICATION OF RANDOMISED

CONTROL METHOD

Abstract

The study1 attempts to investigate the role of education and environmental

concerns in conserving ‘agricultural biodiversity’ of diverse farming systems

in Sri Lanka. It analyses how farmers’ valuation of agricultural biodiversity

alters with education on agricultural biodiversity and environmental

concerns. The Choice Experiment (CE) Method, combined with Randomized

Control Method (RCM) was used to collect data, while models such as

Conditional Logit (CL) and Random Parameter (RP) Logit were used to

analyse the collected data. The results clearly illustrate that farmers’

education level, as well as environmental concerns play a major role in the

conservation of agricultural biodiversity. The overall findings of this study

will help the policy makers to implement relevant policies in reducing

degradation of agricultural biodiversity, which is increasingly posing a

major impediment to agricultural growth, to the environmental protection

and sustainable development.

Keywords: Agricultural Biodiversity, Education, Valuation

P.P.A. Wasantha Athukorala Senior Lecturer, Department of Economics and Statistics, University of Peradeniya,

Sri Lanka

1 Financial assistance received from the Peradeniya University Research Grant (RG/AF

2013/61/A) for this project is greatly appreciated


30

INTRODUCTION

‘Agricultural biodiversity’ is a sub-set of general biodiversity, which

is essential for the global food production, livelihood security,

environmental protection and sustainable agricultural development

(FAO, 2007). The direct and indirect benefits of conserving farm level

biodiversity can be explained on the basis of the Total Economic Value

(TEV) framework. TEV consists of both ‘use’ and ‘non-use’ values.

The benefits obtained by individuals using agricultural biodiversity are

defined as ‘use values.’ ‘Use values’ include, direct, indirect, portfolio

values and option values2 (Primack, 1993; Swanson, 1996; Evenson et

al. 1998). On the other hand, bequest values, altruistic values, existence

values and cultural values of agricultural biodiversity are considered

under ‘non-use values’ (Brown, 1990; Primack, 1993; Evenson et al.

1998).

Some direct benefits that farmers can gain from maintaining a ‘diverse

farming system’ can be smmerised as follows. Firstly, a diverse

farming system minimizes the external risk that farmers often faced.

For example, if a farmer has both crops and livestock, this will

minimize the risk of loss from drought or water shortage. That is,

though crop harvests may be devastated by harsh climatic conditions,

the farmers still can derive an income from livestock. Second, high

levels of agricultural biodiversity provide fresh nutritional foods for

2 Option values can be placed under both use and non-use values. It includes future

direct and indirect use values.


31

the members of their families. Third, a diverse farming system can help

farm families to utilise family labour optimally (Brookfield et al.

2002). For example, crop diversity demands labour in different time

periods and family labour can easily be distributed among different

crops and/or livestock in order to obtain maximum benefits. Fourth,

agricultural biodiversity is found to have positive impacts on the

overall productivity and quality of soil (Karunarathna, 2013). In

addition to providing direct benefits to farmers, agricultural

biodiversity improves ecological processes through regulating climate,

maintaining soil quality, providing protection from erosion, storing

nutrients and breaking down pollution (Di Falco and Chavas, 2009).

Despite all these benefits, the previous experience has shown that

population growth, inequity, inadequate economic policies and

institutional systems have mainly contributed towards the increasing

loss of agricultural biodiversity in the world. Low levels of education

and lack of integrated research on natural ecosystems and their

innumerable components may exaggerate the process, especially in

developing countries.

The overall aim of this study is to evaluate the farmers' preference of

agricultural biodiversity in diverse agricultural systems in Sri Lanka.

Accordingly, the study attempts to investigate the fact that how

farmers’ valuation of agricultural biodiversity would change along

with changes happened in relation to the discrepancies of their

education and environmental concerns. The Choice Experiment (CE)

method (combined with Randomized Control Method) is used to


32

estimate farmers’ preferences for agricultural biodiversity. Models

such as Conditional Logit (CL) and Random Parameter (RP) Logit are

used to analyse the data. The results are then used to estimate the likely

welfare gains under different groups of farmers. The overall findings

of this research will help for implementation of policies to reduce

degradation of agricultural biodiversity in developing countries.

LITERATURE REVIEW

A number of studies have already been undertaken to investigate the

various aspects of agricultural biodiversity (Kontoleon, 2003; Lusk et

al. 2003; Birol, 2006; Ouma et al. 2007; Ruto et al. 2008; Roessler et

al. 2008; Zander and Drucker, 2008; Kassie et al. 2009; Asrat et al.

2009). Brock and Xepapadeas (2003) developed a conceptual

framework for valuing biodiversity from an economic perspective.

This study illustrated that a more diverse system could attain a higher

value, even though the genetic distance of the species in the more

diverse system could be almost zero. Kontoleon (2003) investigated

consumers’ perceptions of ‘Genetically Modified’ (GM) food and

found that consumers across the European Union (EU) were willing to

pay more to obtain information on the GM content in their food

supplies.

Di Falco and Perrings (2003) investigated the impact of providing

financial assistance to farmers in maintaining crop biodiversity in an

uncertain setting. The findings reveal that risk aversion is an important


33

driving force for crop biodiversity conservation. This is because, risk

averse farmers can hedge against uncertainty they face, by allocating

land to different crop species. Van Dusen et al. (2005) carried out an

empirical case study about farmer management of rice genetic

resources in two communities of Nepal. In this study, the decision-

making process of farm households is modelled and estimated in order

to provide information for the design of community-based

conservation programmes. The CE method was used to investigate

farmers’ valuation of agricultural biodiversity of maize varieties, using

414 farm households from three States of Mexico by Birol et al. (2006).

The results revealed that there was a considerable heterogeneity among

the farmers’ preferences for Milpa diversity and GM maize across

within the three States.

Poudel and Johnsen (2009) sought to advance the application of

Contingent Valuation (CV) method to document the economic value

of crop genetic resources, based on farmers’ willingness to pay for

conservation. According to them, landholding size, household size,

education level, socio-economic status, gender of the respondents,

number of crop landraces grown, and the knowledge of biodiversity

influence the willingness to pay for in situ conservation; whereas only

landholding size and household size influence the willingness to pay

for ex-situ conservation. The CE approach was employed to investigate

Ethiopian farmers’ crop variety preferences and to estimate the mean

willingness to pay for each crop variety attributes by Asrat et al. (2009).

They also identified household-specific and institutional factors that


34

governed the preferences. Recently, a Choice Experiment method was

used by Kikulwe et al. (2011) to estimate farmers’ valuation of

agricultural biodiversity in the milpa system, and examined their

interest in cultivating Genetically Modified (GM) maize.

Meanwhile, several studies have used econometric models to identify

the determinants of diversity in livestock and crops in developing or

transitional economies. Some studies, conducted in Peru (potato),

Turkey (wheat), and Mexico (maize), have sought to identify some of

the important factors that positively and negatively affected the

conservation of agricultural biodiversity (Brush et al. 1992; Meng,

1997; Van Dusen, 2000; Smale et al. 2002). However, most of these

studies (Brush et al. 1992; Franks, 1999; Bellon, 2004; Benin et al.

2004) concentrate on diversity within a single crop or animal bread.

When analysing the multiple benefits of the farms under semi-

subsistent rural areas, concentration on variety diversity, mix farming

systems and organic farming are more important than considering a

single crop.

Although these studies identified the importance of conserving

agricultural biodiversity, literature on economic valuation of diverse

farms in semi-subsistence economies are very limited in developing

countries. This is because, assigning monetary values to diverse

farming systems are complicated in subsistence farming systems

(Gauchan, 2004; Diwakar and Johnsen, 2009); therefore, a challenging

area of study. As a result, these studies have only provided limited


35

information on the value of the different attributes of agricultural

biological diversity. This study is expected to fill this void in the

literature by applying CE method to analyse farmer's preference for

different aspects of agricultural biodiversity in Sri Lanka.

CHOICE EXPERIMENT METHOD

CE approach has a theoretical grounding in Lancaster’s Attribute

Theory of Consumer Choice (Lancaster, 1966) and an econometric

basis in models of random utility (McFadden, 1974). Lancaster

proposed that consumers derive satisfaction not from goods

themselves, but from the attributes they provide. To illustrate, the basic

model behind Choice Experiments, it is assumed that farm families

have a utility function of the form:

Where, for any farm family a given level of utility will be associated

with any alternative farm Utility derived from any farm alternatives

depend on the attributes of the farm and the social and economic

characteristics of the farm family , since different families may

receive different levels of utility from these attributes. According to the

‘Random Utility Model’, the utility of a choice comprises of a

systematic (deterministic) component, and an error (random)

,i

.j

ijX

iZ

ijT

)1.........(....................).........,( iijij ZXUU


36

component , which is independent of the deterministic part and

follows a predetermined distribution (Hanemann et al. 1991):

The systematic component can be explained as a function of the

characteristics of the farm and of the social and economic

characteristics of the farm family. Accordingly, Equation 2 can be

expressed as:

Given an error part in the utility function, choices made among

alternative farms will be a function of the probability that the utility

associated with a particular farm option is higher than that for the

other alternative farm. Hence, the probability that farm family will

choose farm over all other options is given by:

Where, .

It is assumed that the relationship between utility and attributes follow

a linear path in the parameters and variables. We further assume that

the error terms are identically and independently distributed with a

Weibull distributio3 (Greene, 1997). These assumptions ensure that the

probability of any particular alternative j being chosen can be

3Weibull distribution is a continuous probability distribution. For further details

about the basic properties of this distribution (1997).

ije

)( j

i

j n

ininijijij eTeTprobP nj

)2.........(..............................ijijij eTU

iiijij eZXTU ),(

http://en.wikipedia.org/wiki/Probability_distribution


37

expressed in terms of logistic distribution. This specification is known

as the ‘CL Model’ (McFadden, 1974; Greene, 1997; Maddala, 1999)

which has the following general form:

The components of Xij are typically called the attribute of the choices.

However, Zi contains characteristics of the individual and is, therefore,

the same for all the choices. Equation 3 is the probabilistic response

function and it shows that, given all other options, the probability of

farmer i selecting the option j type farm. The CL model generates

results for a conditional indirect utility function of the form:

Where, is the alternative specific constant (ASC), that captures the

effects in utility from any attributes, which are not included in choice

specific attributes (Rolfe et al. 2000). The number of farm attributes

considered is m and the number of social and economic characteristics

of the farm family employed to explain the choice of the farm is . The

vectors of coefficients are attached to the vector of attributes and

to a vector of socio-economic factors that influence utility,

respectively. The CE method is consistent with Utility Maximization

and Demand Theory. When parameter estimates are obtained, welfare

k

)(X

)(Z

)3......(..............................)exp(

)exp(

1

''

''

J

j iij

iij

ij

ZX

ZXP

)4........(........... 221122110 kkmmij ZZZXXXT


38

measures can be estimated from the CL model using the following

formula:

Where is the compensating surplus welfare measure, is the

marginal utility of income (generally represented by the coefficient of

the monetary attribute in the CE) and and represent indirect

utility functions of alternative i (with subscript 0 indicating the base

situation and 1 indicate the changed situation) before and after the

change under consideration. For the linear utility index, the marginal

value of change within a single attribute can be represented as a ratio

of coefficients, reducing Equation 5 to 6:

In addition to this, the following Equation 7 can be used to estimate the

consumer surplus in different areas.

)7.........(..........1

cos

ASCCS organicmixcrops

ts

CS

0iT 1iT

)5.....(..............................

)exp(ln)exp(ln 01

i

i

i

i TT

CS

)6....(..............................var_

iablemonetary

attributeW


39

Using Equation 6, the implicit prices (W) for the various farm attributes

can be calculated. These demonstrate the marginal rate of substitution

between cost and the attribute in question. This is the same as the

marginal welfare measure (WTP or WTA) for a change in any of the

attributes. An alternative model specification to the CL model is

random parameter logit (RPL) model which is increasingly becoming

popular in CE studies. The advantage of RPL model is that it accounts

for consumers’ taste heterogeneities and also relaxes the Independence

of Irrelevant Alternatives (IIA) assumption of the CL model. More

importantly, preferences are in fact heterogeneous and accounting for

this heterogeneity, enables estimation of unbiased estimates of

individual preferences and enhances the accuracy and reliability of

estimates of parameters of the model and total welfare (Greene, 1997).

DESIGNING CE STUDY

In this study the most important attributes of farms and their levels

were identified through consultation with experts from the Department

of Agriculture in Sri Lanka; drawing on the results of informal

interviews and workshops with farmers in the study sites, focus group

discussions and a thorough review of previous research in this area in

the country. After identifying the attributes for the experiment, we

assigned values or levels to each attribute. Selected attributes for this

study, their definitions and the levels are given in Table 1.


40

Table 1: Identified Attributes of Agricultural Biodiversity in the

Country

Attributes Definition Levels

Number of

species

This is measured by the total number of crops and

animal species that are cultivated in the farm, in a

given season

1, 4,

7, 10

Mixed

farming

system

This attribute investigates whether a farmer prefers

an integrated crop and livestock production system

to a system that is specialised in crops or livestock.

Yes,

No

Organic

production

This attribute investigates whether a farmer prefers

organic methods of production to a system using

chemical fertiliser and pesticides

Yes,

No

Estimated

costs in

terms of

additional

labour

days

This is defined as a percentage of additional labour

requirements under different policy options. This is

the monetary attributes of the CE study.

5 %,

10%

15%

Note: i. Upper and lower bound of the crop species diversity and additional labour

requirements are estimated using pilot survey information.

Table 1 shows the attributes that we used in the CE study. The first

three attributes reflect the various attributes of a diverse farming

system, found in the farms in Sri Lanka. The last factor is the monetary

attribute in terms of additional labour costs4 that farmers have under

different policy options. As compared to willingness to pay (WTP),

4This indirect measure is preferred over a direct monetary attribute due to three

reasons. First, most of the outputs and functions of the farms that are considered in

this study are not traded in the markets, but consumed by the farm families

themselves. Therefore, they are not likely to be familiar with a direct monetary

measure. Second, the proxy monetary attribute can easily be converted into actual

monetary units by using secondary data on labour costs. Third, it is obvious that a

diverse farming system requires more labour than a specialised farming system.

Therefore, monetary attribute that measure the additional labour costs that farmers

have to allocate for receiving the benefits of agricultural biodiversity may work as a

good proxy for measuring welfare changes in the community in this study.


41

willingness to accept is measured as a benefit rather than a cost

(Freeman, 2003). In order to estimate this benefit, a monetary attribute

in terms of additional labour costs that farmers are willing to offer is

included. The size of the hypothetical farm is fixed at one acre in area

in each case (this is the average farm size in study area). A large

number of different types of farms (combinations of attributes) could

be constructed from this number of attributes and levels. However, in

this study we used 16 choice options and they were randomly blocked

into 4 different versions (each has four options). Using the Dptimal

procedure in Engine, an experimental design was undertaken to

recover 16 pair wise comparisons of farm profiles. A sample choice set

presented to respondents is given below.

Assuming that the following farms were the ONLY choices you

have, which one would you prefer to cultivate?

Farm Characteristics Farm (A) Farm (B)

Neither

farm (A)

nor farm

(B)

Number of species 1 10

Mixed farming system Yes No

Organic production No No

Estimated costs in terms

of additional labour days

5 10

I prefer to cultivate Farm (A)…...................

Farm (B)…....................

Neither Farm .......….…

(Please pick one option)


42

SURVEY METHOD, DATA CORDING AND ESTIMATION

PROCEDURE

This study employs Randomized Control Method (RCM) to collect the

data. Firstly, 21 villages from Ampara district were selected

purposively. Then, the villages were randomly divided into three

groups: one control groups and two treatment groups. Accordingly,

randomisation to select households into different groups (two

treatment and the control) was done at the village level and ‘village’ is

the unit of randomisation in this study. Then, using the farmers' list, 20

farmers were randomly selected from each village. The survey

involved several steps. First, households for the survey from the district

were selected. Second, based line survey was carried out covering all

three groups. The main purpose of this survey is to understand their

education and environmental attitudes on biodiversity and test whether

there is any significant spill over effect among the respondents. This

survey includes a set of simple questions related to agricultural

biodiversity and environment concerns5. Third, one treatment group

was educated using a formal educational programme (specially

designed for this study) on agricultural biodiversity, while other

treatment group was provided with information in order to improve

their environmental concerns. Control group was not provided with

5 During the final survey, the same questions were asked. By comparing the

information of those two surveys, spillover effects can be judged. For example, we

can check whether there is a significant change of the knowledge on

biodiversity/environmental concerns of the control group.


43

any information. Finally, CE survey6 covers all groups in the district.

After one month of the formal educational programme, the final survey

was conducted using structured questionnaire which was developed

based on the results from six focus group discussions and a pre-test. A

pre-test for 15 respondents was undertaken in February 2013. On the

basis of the pre-test, only minor modifications to the questionnaire

were required. In the questionnaire, respondents were said that the

development of the Choice Experiment questionnaire was based on

focus group studies. Six focus group discussions were conducted for

both potential respondents (3) and for agricultural officers (3) to ensure

that inputs for choice sets were correctly specified.

Education programme consisted with two steps. Firstly, we met the

respondents of two control groups and explained the importance of

agricultural biodiversity or environmental protection individually.

Secondly, we provided leaflets (but not keep with them; they could

read it in front of us or we can read it for them) showing the importance

of maintaining diverse farming systems in their farms or the

importance of protecting environment. Final survey was carried out by

administering a questionnaire through a face-to-face interview with the

Head or any other working member of the households. The final survey

covered 420 households.

6 Before the interview it was confirmed whether the respondents were generally

those responsible for farm production decision making.


44

Data coding is one of the important parts of the Choice Experiment

Model. During this study, the data was coded according to the levels

of the attributes. Attributes with 2 levels entered the utility function as

binary variables that were effects coded (Louviere et al. 2000). Crops

diversity variable is used as a continuous variable. Consequently, the

crop species diversity attribute took levels 1, 4, 7 and 10. Estimated

costs in terms of additional labour requirement were transformed into

monetary values when estimating the models. The percentage values

of additional labour requirements were given as 5, 10 and 15. On

average, farm families need approximately 31 labour days per month

for their one acre farm cultivation. Daily average wage rate per person

was Rs.5007. Accordingly, value of the cost of accepting alternative

farms can be expected as Rs. 2,500, Rs. 5,000 and Rs. 7,500 for the

three levels respectively. In this way, the levels used for labour

requirement variables were entered in a cardinal-linear form. The

attributes for the ‘neither farm’ option were coded with zero values for

all attributes. The alternative specific constants were equalled to 1,

when either farm A or B was chosen and to 0 when the ‘neither farm’

alternative was chosen. Choice data were converted from wide to long

format with a programme coded in LIMDEP 9.0 NLOGIT 4.0. This

data conversion step was necessary in estimating models with multiple

responses from each respondent; a format similar to panel data.

7 This varies between Rs. 600 and Rs. 400 depending on various factors (gender,

period and area). For example, men wage rate is slightly higher than female. Wage

rate in harvesting period is greater than other period.


45

First, we estimated the CLM. The IIA property of this model is tested

using a procedure suggested by Hausman and McFadden (1984). This

test involves constructing a likelihood ratio test around different

versions of the model, where choice alternatives are excluded. If IIA

holds, the model estimated on all choices (the full choice set) should

be the same as that estimated for a sub-set of alternatives (Bateman et

al. 2003). It is found that the IIA conditions have not violated any of

the cases. We then included social and economic characteristics as

interaction terms, and test whether there is an improvement of our

result. It was found to not have any significant improvement in

including any social-economic characteristics as the interaction term.

As the next step of the analysis, Random Parameter Logit (RPL) model

was used in order to take into account the preference heterogeneity. We

estimated the basic RPL model, which includes only attributes as well

as extended RPL model that includes some socio-economic variables.

When comparing with RPL results with CL results, it was found that

basic CL results were better in terms of overall fit of the model and

number of significant variables. Therefore, the result of the basic CL

model was used to simulate welfare change of the society, when

changing different attributes and their level.

RESULTS AND DISCUSSION

The descriptive statistics of the samples show the similar

characteristics of all three groups. The mean values of age were 38, 40

years for two treatment groups and 37 years for the control group. The


46

average number of persons in the household was 5, 4 and 5 while the

average education levels were 10, 9 and 9, respectively. Although

agriculture was the dominant source of household income, monthly

income from non-farm activities was approximately estimated as Rs.

1,350, Rs. 1,300 and Rs. 1,425 per household; which accounted for

almost 5 percent of the total household income. The mean labour usage

per season was 74 man-days for three samples. This is expected, given

the tedious labour intensity for all agricultural work in semi-

subsistence economy. Rice was cultivated by almost all households

followed by various types of vegetables and cash crops. The maximum

number of crop varieties cultivated by any household was seven. More

details of the descriptive are given in Table 2. The comparison of

descriptive statistics clearly indicate that there is no significant

difference among different groups.

Before estimating any model, we explored a variety of different

specifications of the utility functions to identify the best specification

of the data. These tests include both formal statistical tests and informal

judgments about the signs, magnitudes, or relative magnitudes of

parameters based on our knowledge about the underlying behavioural

relationships that influence different choice of farms.


47

Table 2: Descriptive Statistics of the Variables Used in the Study

Variable Average Maximum Minimum

Age 38.33 62.00 18.00

Household size 4.66 7.00 2.00

Education level 9.33 12.00 4.00

Farm income

(Rs. per month)

26500.00 62450.00 6540.00

Off- farm

income

(Rs. per month)

1358.30 6500.00 0.00

Labour days

(per month)

24.66 45.00 12.00

Note: As hypothetical scenarios are given in choice experiment, descriptive statistics

of those variables are not recorded.

As a formal statistical process, different model specifications were

compared according to higher log-likelihood value criterion in this

study. Most appropriate specification was found to be the model with

the linear version of the four attributes of the study. Accordingly, CL

model is specified, so that the probability of selecting a particular

alternative is a function of attributes of the alternatives and of the

alternative specific constant. Indirect utility received by the farm

attributes take the form:

)8.....().........()()()( cos43210 torganicmixspeciesij XXXXT

Where, β0 refers to the alternative specific constant and β1-4 refers to

the vector of coefficients associated with the vector of attributes

describing farm characteristics. The results of the estimated basic CL

model for the separate samples are presented in Table 3. All attributes


48

except the organic farms in the control group in the models were

statistically significant at conventional levels, and their signs were as

expected.

Table 3: Regression Results of the CL Model

Variables Model 1 Model 2 Model 3

Education

(biodiversity)

Education

(Environment)

Control group

ASC 2.32

(0.009)*

3.028

(0.005)*

2.984

(0.029)**

Crop

diversity

0.081

(0.000)*

0.119

(0.002)*

0.011

(0.076)***

Mix system 0.229

(0.001)*

0.446

(0.001)*

0.022

(0.008)*

Organic

farms

0.196

(0.041)**

0.477

(0.020)**

0.034

(0.341)

Costs -3.1E-04

(0.000)*

-2.5E-04

(0.000)*

-2.1E-04

(0.000)*

LR chi2(5) 956.24 1425.14 971.44

Prob > chi2 0.000 0.000 0.000

Pseudo R2 0.228 0.321 0.120

N 1680 1680 1680

Note: i. P values are shown within parentheses

ii. *denotes significant at 1% level while ** and *** indicates

significant variables at 5% and 10% level respectively

All attributes in the first and second models were statistically

significant at conventional levels (5 percent), and their signs were as

expected. The overall fit of the model as measured by McFadden’s R2

was also good by conventional standards used to describe probabilistic

discrete choice models (Ben-Akiva and Lerman, 1985). When

analysing the results of these two models, it is clear that all of the farm

attributes are statistically significant at 5 percent level, implying that


49

any single attribute increases the probability that a farm is selected;

whereas different socio-economics variables and preferences remain

equal. Since the underlying samples is statistically significant

(P<0.05), these parameters represent preference estimates of farm

families for farms attributes among two treatment groups. Third model

shows the situation of the control group. For this group, organic farm

variable is not significant (P>0.01) even under ten percent level while

all other attributes were significant at five or ten percent level.

Relatively, significance of the attributes of this model is less when

compared to other two models. The results clearly show that the

education has some impact on farmers’ valuation of agricultural

biodiversity in study area.

As the next step of the analysis, the IIA property of all models is tested

using a procedure suggested by Hausman and McFadden (1984) and

contained within NLOGIT 4.0. This test involves constructing a

likelihood ratio test around different versions of the model where

choice alternatives are excluded. If IIA holds, then the model estimated

on all choices (the full choice set) should be the same as that estimated

for a sub-set of alternatives (Bateman et al. 2003). It was found that the

IIA property is not violated implying that the conditional logit

estimates do not hold any bias that could have resulted from inclusion

of the ‘neither’ option. However, CL model assumes homogeneous

preferences across farm families in each district, though references

across families are in fact heterogeneous. Accounting for this

heterogeneity enables estimation of unbiased estimates of individual


50

preferences and enhances the accuracy and reliability of parameter

estimates (Rolfe et al., 2000). Interaction of individual-specific social

and economic characteristics with choice specific attributes or with

ASC of the indirect utility function is a common solution to dealing

with the heterogeneity. However, the main problem with this method

is multi-collinearity, which occurs when too many interactions are

included in the estimation. In this context, the model needs to be tested

down, using the higher log-likelihood criteria (Bateman et al., 2003;

Birol, 2004). Therefore, as the next step of the analysis, CL model is

estimated using three socio-economic variables as interaction terms8.

They are; age of the respondent (age), education level of the respondent

(education) and household size (hhs). Accordingly, indirect utility

received by the farm attributes and interaction with socio-economic

characteristics can be re-specified as Equation 9:

)()()()()()( 21cos43210 eduspeciesagespeciestorganicmixspeciesij ZXZXXXXXT

)()()()()( 76543 ageorganichhsmixedumixagemixhhsspecies ZXZXZXZXZX

)9(................................................................................).........(

)()()()(

cos12

cos11cos1098

hhst

edutagethhsorganicedusorganic

ZX

ZXZXZXZX

As it shown in equation 9, the total number of coefficients in the full

model is 16. We tested various interactions of the four farm attributes

with the household-level including other characteristics. An initial run

8 Different socio-economic variables were tried and finally those three were found

to be more appropriate.


51

of the model with all interaction terms reveal that a large number of

variables are insignificant for all three models. Then we estimated the

correlation matrix and it was revealed that there was a higher level of

correlation and multi-collinearity among these household-level

variables. Estimation of variance inflation factor further provided the

evidence about the higher correlation among household level variables.

To address this limitation (except age, education and household size)

all other independent variables were eliminated based on variance

inflation factors, which were calculated by running ordinary least

square regressions between each independent variable9. The estimated

results of the final models are reported in Table 4.

This specification of the model was not significantly different from the

previous specification. In particular, the model did not reveal a

significant higher level of parametric fit, compared with the first

model. Most of the interaction terms of all three models are not

significant. Further, including the interaction terms has reduced the

significance of some of the attributes of the models.

Therefore, it can be concluded that the improvement in model fit was

not significant.

9 Those independent variables for which VIFj >0.6 indicate clear evidence that the

estimation of the characteristic is being affected by multicollinearity (Maddala,

2000).


52

Table 4: CL Model Including Attributes and Socioeconomic

Variables

Note: i. *denotes significant at 1% level while ** and *** indicates significant

variables at 5% and 10 % level.

ii. P values are given in parenthesis.


Education

(biodiversity)

Education

(Environment)

Control group

ASC 1.42 (0.016)** 2.317 (0.004)* 2.984(0.019)**

Crop diversity 0.136 (0.002)* 0.362 (0.008)* 0.019 (0.169)

Mix system 0.124 (0.016)** 0.219 (0.021)** 0.063 (0.192)

Organic farms 0.116 (0.042)** 0.192 (0.000)* 0.044(0.041)**

Costs -2.2E-4(0.000)* -2.9E-05(0.000)* -1.6E-04(0.000)*

Crops_age 0.026 (0.231) 0.051 (0.090)*** 0.002 (0.201)

Crops_edu 0.002 (0.041)** 0.0160 (0.000)* 0.002 (0.131)

Crops_hhs 0.002 (0.620) 0.001(0.000)* 0.017 (0.870)

Mix_age 0.107 (0.821) 0.203 (0.141) 0.007 (0.121)

Mix_edu 0.001(0.060)*** 1.425(0.059)*** 0.074 (0.083)***

Mix_hhs 0.012(0.061)*** 1.224 (0.587) 0.704(0.065)**

Organic_age 0.003 (0.506) 0.029 (0.601) 0.043 (0.801)

Organic_edu 0.043 (0.018)** 1.098 (0.017)* 0.247(0.060)***

Organic_hhs 0.01 (0.255) 0.051(0.401) 1.248 (0.294)

Costs_age -0.073 (0.459) -1.13 (0.410) -0.233(0.094)***

Costs_edu -0.305 (0.129) 0.451 (0.016)** 1.324 (0.183)

Costs_hhs -0.031(0.149) -0.130 (0.40) -0.203(0.840)

LR chi2 1276.23 1825.88 1141.64

Prob > chi2 0.000 0.000 0.000

Pseudo R2 0.229 0.342 0.140

N 1680 1680 1680


53

The Hausman-McFadden test also revealed that CL model without

interactions is a better fit for the data, than the CL model with

interaction. Among the significant interactions, households with a

higher education level in all samples had a higher preference for

organic and mix farming practices. The overall model is significant at

the one percent level. Compared to basic CL model, the explanatory

power of the model has not changed significantly.

An alternative method to account for preference heterogeneity is the

use of RPL model10. We next estimate the results using RPL model to

investigate the fact that whether there is an observable improvement of

the results. Running the RPL model requires an assumption to be made

about the distribution of preferences for each attribute. The main

candidate distributions are normal and log normal. The former allows

preferences to range between positive and negative for a given

attribute, the latter restricts the range to being of one sign only. Further,

treating preference parameters as random variables require estimation

by simulated maximum likelihood. This means that the maximum

likelihood algorithm searches for a solution by simulating m draws

from distributions with given means and standard deviations.

Probabilities can be calculated by integrating the joint simulated

distribution. In this study the RPL model was estimated using NLOGIT

10 RPL model is one of the fully flexible versions of the discrete choice models

because its unobserved utility is not limited to the normal distribution. It

decomposes the random parts of utility into two parts. One has the independent,

identical type 1 extreme value distribution, and the other representing individual

tastes can be any distribution. It is also characterized by accommodating

heterogeneity as a continuous function of the parameters.


54

4.0. All the parameters were specified to be independently normally

distributed and distribution simulations were based on 500 draws. The

results of the RPL estimations for the separate districts are reported in

Table 5

Table 5: Regression Results of the RPL Model for Separate

Districts and Pool Data


Education

(biodiversity)

Education

(Environment)

Control group

ASC 1.838 (0.042)** 1.343 (0.007)* 2.428 (0.013)**

Crop

diversity

0.0640 (0.009)* 0.126(0.009)* 0.015 (0.032)**

Mix system 0.178 (0.040)** 0.157(0.001)* 0.135(0.081)***

Organic

farms

0.147 (0.044)** 0.192(0.036)** 0.184 (0.145)

Costs -2.0E-0.4(0.000)* -2.3E-

05(0.000)*

-1.4E-04(0.000)*

Log

likelihood

-1722.65 -2943.43 -1647.32

Simulation 500 500 500

ρ2 0.178 0.198 0.161

N 1680 1680 1680

Notes: i. P values are shown in brackets.

ii. *denotes significant at 1% level while ** and *** indicates significant

variables at 5% and 10 % level respectively.

The results of the RPL model are quite similar in sign and magnitude

to the CL model, where preferences are assumed to be homogenous.

The crop diversity coefficient for the standard CL model is 0.081,

whereas it is 0.064 for the RPL for Model 1. In this model coefficients

of the mix farming systems are 0.229 and 0.178 for CL model and RPL

model respectively. The CL model contains all positive and significant

choice attributes, except organic farm in model three which is similar


55

to the RPL results. The major difference between the two models is

that most coefficients of the CL models are highly significant, while

RP models are relatively less significant. The Swait Louviere Log

Likelihood ratio test results of the test cannot reject the null hypothesis

that the RPL model and CL model estimates are equal. Hence no

improvement in the model fit can be achieved with the use of a RPL

model. Accordingly, it can be concluded that the CL model is sufficient

for analysis of the data set presented in this study. Therefore, the results

of the CL model reported in Table 2 can be used to calculate the value

assigned by the farm families to farm attributes. Point estimates of the

WTA, a change in one of the attributes in the choice sets can be found

by estimating implicit prices. Implicit prices are the marginal rates of

substitution between the attribute of interest and the monetary attribute.

This is equal to the ratio of the coefficient of one of the non-monetary

attributes and the monetary attributes. Equation 6 is used to estimate

the implicit prices for each attribute. Estimates of implicit prices for

each of the non-monetary attributes in the choice sets are reported in

Table 6.


56

Table 6: Implicit Price Estimates for Attributes

Variables Education

(biodiversity)

Rs.

Education

(Environment)

Rs.

Control

group

Rs.

Crop

diversity

261 476 86

Mix

system

739 1784 105

Organic

farms

632 1908 162

Note: all implicit prices are estimated using the result of the basic CLM.

These estimates indicate that, for example, farmers’ valuation of the

additional benefits that farmers could obtain in increasing crop

diversity by one per month is Rs. 86, 261 and 476 in control group,

educated on biodiversity group and educated on environment group

respectively. It is clear that farmers of ‘educated on environment’

group have placed relatively high values on all attributes. Interestingly,

the highest valuation is given for organic farming when comparing

with other attributes. This is expected as most farmers in these districts

use their farm products for their own consumption.

These estimates of implicit prices are based on a ceteris paribus

assumption, where we assume that all other parameters are held

constant except the attribute for which the implicit price is being

calculated. Implicit prices, however, do not provide estimates of

compensating surplus. Estimating the overall WTA for a change from

the current situation requires more substantial calculations. This is

because, the attributes in the choice sets do not capture all of the


57

reasons why respondents might choose to increase agricultural

biodiversity. To estimate overall WTA, it is necessary to include the

alternative specific constant. As discussed earlier, the alternative

specific constant captures systematic but unobserved information

about why respondents chose a particular option (unrelated to the

choice set attributes). Therefore, as the final step of the analysis,

consumer surpluses for different samples are estimated using Equation

7. Then changes of the CS relative to the control group are estimated.

These changes are given in Table 7.

Table 7: Changes of the CS under Each Scenario

Crops

diversi

ty

Mix Organic Education

(Biodiversity)

CS / Rs.

Education

(Environment)

CS / Rs.

Status

quo

1 0 0 - -

Scenario 1 4 1 0 2,434 4,522

Scenario 2 7 1 1 3,127 4,786

Scenario 3 10 1 1 3,434 5,245 Note: After estimating the CS per farmer under each scenario, the deviation from the

status quo were calculated. Then the changes of those two groups from the

control group were estimated.

To illustrate this process, estimates are provided for three alternative

scenarios. The status quo situation is identified as a farm with one crop

variety and no mixed farming or organic farms. We changed these

characteristics for the rest of the profile gradually and estimated change

of the CS under each profile. Estimated change of CS in each sample

is given in Table 7. The CS values indicate that the value attached to

deviation from control group to educated on biodiversity group were


58

Rs. 2,434, 3,127 and 3,434 respectively. Those values for educated on

environment group were Rs. 4,522, 4,786 and 5,245 respectively.

These are, the average benefits per household which can be obtained

by increasing crop diversity, adopting mix farming practice and

introducing organic farming practice in their farms. This shows that

farmer welfare could be easily increased by shifting farming practice

to more diverse systems in rural areas in Sri Lanka.

CONCLUSIONS AND POLICY IMPLICATIONS

This research is one of the first attempts to use choice modelling to

investigate farmers’ preference for different attributes of agricultural

biodiversity that can be seen in diverse farms in Sri Lanka. The study

applied the choice modelling approach with RCM to identify question

that whether farmers preferences on different attributes of agricultural

biodiversity are affected by different types of education or not? The

result of this study shows that agriculture based families in the study

area have strong preferences to increase agricultural biodiversity. It is

clear that all diversity components are valued highly by all types of

households in the study area.

In general, the findings of the Choice Experiment support the

assumption that diverse farms and their multiple attributes contribute

positively and significantly to the utility of farm families in Sri Lanka.

Furthermore, their valuation is highly affected by their education on

biodiversity and environmental attitudes. The present study illustrates


59

the importance of education in determining farmers’ valuation of

agricultural biodiversity. Therefore, designing formal and informal

education programmes will improve farmers’ abilities to conserve

agricultural biodiversity in the country. Moreover, a further initiative

can be taken to strengthen the capacity of farmers through farmer-

centred training workshops geared towards conserving agricultural

biodiversity. This could be done in a collaborative manner involving

the government, district assemblies and NGOs. The government also

needs to intensify its agricultural extension service programmes by

training and deploying qualified extension officers. The officers, in

turn, should intensify farmer education on agricultural biodiversity.

The overall findings of this research will help to implement policies to

reduce degradation of agricultural biodiversity that is increasingly

posing a major impediment to agricultural growth, environmental

protection and sustainable development. Further, the research findings

will contribute to the sustainable use of agricultural biodiversity to

improve farmers’ well-being and achieve an environmentally friendly

farming system. It also helps to increase awareness and generates

support for investment in conservation and development of agricultural

biodiversity. Moreover, it will provide an opportunity to make

necessary policies that provide incentives to protect biodiversity at

farm level that generate regional as well as global benefits in the future.


60

REFERENCES

Bateman, I. J., Carson, R. T., Day, B. W., Hanemann, M., Hanley, N.,

Hett, T., Jones-Lee, M., Loomes, G., Mourato, S.,

Ozdemiroglu, E., Pearce, D. W., R. Sugden and S.

Swanson. 2003. Guidelines for the Use of Stated

Preference Techniques for the Valuation of Preferences for

Non-Market Goods. Cheltenham: Edward Elgar.

Bellon, M. R. 2004. Conceptualizing interventions to support on-farm

genetic resource conservation.World Development, 32(1):

159-172.

Ben-Akiva, M. and S. R. Lerman. 1985. Discrete Choice Analysis.

Theory and Application to Travel Demand. The MIT Press,

Cambridge.

Benin, S. B., Smale, M., Gebremedhin, B., J. Pender and S. Ehui. 2004.

The economic determinants of cereal crop diversity on

farms in the Ethiopian highlands. Agricultural Economics,

31(2): 197-208.

Birol, E., M. Smale and A. Gyovai. 2006. Using a choice experiment

to estimate farmers’ valuation of agricultural biodiversity

on Hungarian small farms. Environmental and Resource

Economics, 34(4): 439-469.

Brock, W. A. and A. Xepapadeas. 2003. Valuing biodiversity from an

economic perspective: a unified economic, ecological and

genetic approach. American Economic Review, 93(5):

597-614

Brookfield, H., Padoch, C., H. Parsons and M. Stocking. 2002.

Cultivating Biodiversity: Understanding, Analysing and

Using Agricultural Diversity. The United Nations

University, ITDG Publishing.


61

Brush, S. B., J. E. Taylor and M. Bellon. 1992. Technology adoption

and biological diversity in Andean potato agriculture.

Journal of Development Economics, 39(2): 365-387

Brown, G. M. 1990. Valuation of Genetic Resources. In: The

Preservation and Valuation of Biological Resources,

Orians, G.H., Brown, G.M., W. E. Kunin and J. E.

Swierzbinski. (Eds.), University of Washington Press, pp:

203-228.

Di Falco, S. and C. Perrings. 2003. Crop genetic diversity, productivity

and stability of agroecosystems. A theoretical and empirical

investigation. Scottish Journal of Political Economy, 50(2):

207-216.

Di Falco, S. and J. P. Chavas. 2009. On crop biodiversity, risk exposure

and food security in the highlands of Ethiopia. American

Journal of Agricultural Economics, 91(3): 599-611.

Diwakar, P. A. and F. H. Johnsen. 2009. Valuation of crop genetic

resources in Kaski, Nepal: farmers' willingness to pay for

rice landraces conservation. Journal of Environmental

Management, 90(1): 483-491

Evenson, R. E., D. Gollin and V. Santaniello. 1998. Agricultural

Values of Plant Genetic Resources. CAB International,

Wallingford.

FAO. 2007. The State of Food and Agriculture 2007. Part I: Paying

farmers for environmental services. Rome.

Franks, J. R. 1999. In situ conservation of plant genetic resources for

food and agriculture: a UK perspective. Land Use Policy,

16(2): 81-91.

Freeman III, A. M. 2003. The Measurement of Environmental and

Resource Values: Theory and Methods, (2nd Eds.),

Resources for the Future, Washington, D.C.

http://www.sciencedirect.com/science/article/pii/S0301479708000261?_fmt=high&_rdoc=1&_origin=article&_zone=cited_by_hover&md5=846c7abf539849927a8b23ce63b04df7




62

Gauchan, D. 2004. Conserving Crop Genetic Resources On-Farm: The

Case of Rice in Nepal, Unpublished Doctoral Dissertation,

University of Birmingham, UK.

Greene, W. H. 1997. Econometric Analysis. (3rd Eds.), New Jersey:

Prentice Hall.

Greene, W. H. 2000. Econometric Analysis. (4th Eds.), New Jersey:

Prentice Hall.

Hanemann, W. M., J. Loomis and B. Kanninen. 1991. Statistical

efficiency of double bounded dichotomous choice

contingent valuation. American Journal of Agricultural

Economics, 73: 1255-1263.

Hanley, N., MacMillan, D., Wright, R. E., Bullock, C., Simpson, I., D.

Parsisson and B. Crabtree. 1998. Contingent valuation

versus choice experiments: estimating the benefits of

environmentally sensitive areas in Scotland. Journal of

Agricultural Economics, 49(1): 1-15.

Hausman, J. and D. McFadden. 1984. Specification tests for the

multinomial logit Model. Econometrica, 52(5): 1219-1240.

Kassie, G. T., A. Abdulai and C. Wollny. 2009. Valuing traits

indigenous cows in central Ethiopia. Journal of


Kikulwe, E. M., Birol, E., J. Wesseler and J. Falck-Zepeda. 2011. A

latent class approach to investigating demand for

genetically modified banana in Uganda.Agricultural

Economics, 42 (95): 547-560.

Kontoleon, A. and M. Yabe. 2003. Assessing the impacts of alternative

‘opt-out’ formats in choice experiment studies: consumer

preferences for genetically modified content and

production information in food. Journal of Agricultural

Policy and Resources, 5: 1-43.


63

Lancaster, K. 1966. A new approach to consumer theory. Journal of

Political Economy, 74(2): 132-157.

Lusk, J. L., J. Roosen and J. A. Fox. 2003. Demand for beef from cattle

administered growth hormones or fed genetically modified

corn: a comparison of consumers in France, Germany, the

United Kingdom and the United States. American Journal

of Agricultural Economic, 85(1): 16-29

Maddala, G. S. 1999. Limited Dependent and Qualitative Variables in

Econometrics. Cambridge University Press, Cambridge.

McFadden, D. 1974. Conditional Logit Analysis of Qualitative Choice

Behaviour. In: Frontiers in Econometrics, Zarembka, P.

(Eds.), New York: Academic Press.

McFadden, D. and K. Train. 2000. Mixed MNL models of discrete

response. Journal of Applied Economics, 15(5): 447-470.

Meng, E. C. H. 1997. Land Allocation Decisions and In Situ

Conservation of Crop Genetic Resources: The Case of

Wheat Landraces in Turkey. Unpublished Doctoral

Dissertation, University of California at Davis, California,

USA.

Ouma, E., A. Abdulai and A. Drucker. 2007. Measuring heterogeneous

preferences for cattle traits among cattle-keeping

households in East Africa. American Journal of


Poudel, D. and F. H. Johnsen. 2009. Valuation of crop genetic

resources in Kaski, Nepal: farmers’ willingness to pay for

rice landraces conservation. Journal of Environmental

Management, 90(1): 483-491.

Primack, R. 1993. Essential of conservation biology, Sinauer Sinauer

Associates,Inc. Publishers: Sunderland, MA.


64

Roesslera, R., Drucker, A. G., Scarpad, R., Markemanna, A., Lemkea,

U., L.T. Thuye and A. V. Záratea. 2008. Using choice

experiments to assess smallholder farmers' preferences for

pig breeding traits in different production systems in North-

West Vietnam. Ecological Economics, 66(1): 184-192.

Rolfe, J., J. J. Bennett and J. Louviere. 2000. Choice modelling and its

potential application to tropical rainforest preservation.

Ecological Economics, 35(2): 289-302.

Ruto, E., G. Garrod and R. Scarpa. 2008. Valuing animal genetic

resources: a choice modeling application to indigenous

cattle in Kenya. Agricultural Economics, 38(1): 89-98.

Smale, M., Meng, E., J. P. Brennan and R. Hu. 2002. Determinants of

spatial diversity in modern wheat: Examples from Australia

and China. Agricultural Economics, 28(1):13-26.

Swanson, T. 1996. Global values of biological diversity: the public

interest in the conservation of plant genetic resources for

agriculture. Plant Genetic Resources Newsletter, 105: 1-7.

Van Dusen, E. 2000. In Situ Conservation of Crop Genetic Resources

in Mexican Milpa Systems. Unpublished Doctoral

Dissertation. University of California, Davis, USA

Van Dusen, E. 2005. Understanding the Factors Driving on Farm Crop

Genetic Diversity: Empirical Evidence from Mexico. In:

Agricultural Biodiversity and Biotechnology in Economic

Development, Cooper, J., L. M. Lipper and D. Zilberman

(Eds.), Springer: US, pp. 127-145.

http://www.springerlink.com/content/u533l7/?p=ac39ea80781349f3baa864a76691b617&pi=0

http://www.springerlink.com/content/u533l7/?p=ac39ea80781349f3baa864a76691b617&pi=0


Economic Research

Volume 3 (1)

June 2015: 65-78

Sri Lanka Forum of


SLJER

K. Saranyah

T.Mahendran

________________________________________________

65

STANDARDIZATION AND

CHARACTERIZATION OF

VALUE ADDED WATERMELON JUICE

(Citrullus lanatus)

READY-TO-SERVE BEVERAGE

Abstract

An experiment was conducted to study the feasibility of blending pineapple

and watermelon juice in different ratio for preparation of blended RTS

beverage. The formulated RTS beverages were evaluated for various

chemical, organoleptic and microbial qualities. The titrable acidity, ascorbic

acid, total soluble solids (TSS) and total sugars of freshly made RTS

beverages increased while pH decreased with the increased concentration of

pineapple juice from 10-30%. Tukey’s test results revealed that, the mean

scores for all assessed organoleptic characters varying significantly

(pH0.05) in the freshly made pineapple blend watermelon RTS beverages.

The formulated beverage with 80:20 watermelon and pineapple juice was

found to be superior in quality and could be stored at 30±1°C for a minimum

period of three months without any significant changes in quality attributes.

Keywords: Pineapple, quality characteristics, Ready-To-Serve beverage,

watermelon.

___________________________________________________

K. Saranyah Department of Agricultural Chemistry, Faculty of Agriculture,

Eastern University, Sri Lanka

T. Mahendran Department of Agricultural Chemistry, Faculty of Agriculture,

Eastern University, Sri Lanka


66

INTRODUCTION

Fruit and Vegetable sector has much potential to contribute to increase

the level of national income, export revenue, generate new

employment opportunities, increase farm income and enhance the

nutrition and health of the people. The potential for cultivating fruit

and vegetable crops for the domestic and export markets is high. Sri

Lanka’s per capita consumption of fruits and vegetables remains far

below the required average daily intake. According to Medical

Research Institute (2011), the daily per capita requirement of fruits for

a balanced diet should be 30‐40g (edible portion), which is

approximately equivalent to 25‐40 kg fresh fruit per head per year.

Consuming fruits and vegetables of all kinds has long been associated

with a reduced risk of many lifestyle-related health conditions.

Watermelon (Citrullus lanatus) is an important cucurbitaceous crop,

whose fruits serve as a dessert to the common people in summer. These

fruits are available in May to June in the markets throughout Sri Lanka.

Watermelon is consumed as a fresh fruit and for its dry seeds. The

watermelon fruit contains 93% water, with small amounts of protein,

fat, minerals, and vitamins. The major nutritional components of the

fruits are carbohydrates, especially sugars (6.4g/100g), vitamin A (590

IU), and lycopene (4,100μg/100g, range 2,300–7,200), an anti-

carcinogenic compound found in red flesh watermelon (USDA

Nutrient Database, 2009). Only a few studies have dealt with drying of

watermelon, mainly with osmotic dehydration of the pulp (Falade et

al., 2007) and spray-drying of the juice (Quek et al., 2007).


67

Pineapple (Ananas comusus) belongs to the family of Bromeliaceae.

The fibrous flesh of pineapple is yellow in colour and has a vibrant

tropical flavour that balances the tastes of sweet and tart. It is an

excellent source of vitamin C which is required for the collagen

synthesis in the body. It is also available in the same season from May

to June of which watermelon are available in Sri Lanka.

Juice blending is one of the best methods to improve the nutritional

quality of the juice. It can improve the vitamin and mineral content

depending on the kind and quality of fruits and vegetables used (De

Carvalho et al., 2007). Apart from nutritional quality improvement,

blended juice can be improved in its effects among the variables, thus

it cannot depict the net effects of various parameters on the reaction

rate. Moreover, one could think of a new product development through

blending in the form of a natural health drink, which may also be served

as an appetizer. The watermelon fruit juice blend that could lower the

risk of cardiovascular disease has been created by Cyril et al. (2009).

Polyphenols are found in fruits as diverse in red watermelon. Their

beneficial effect on health is well known, with a diet rich in

polyphenols correlating with a reduced risk of cardiovascular disease.

According to the Sri Lanka Standard Institute specifications, RTS fruit

beverage should not exceed its total acidity more than 1% (as

anhydrous citric acid). The fruit and sugar content should not be less

than 5% by mass and preservatives should not exceed the limit of

Sulphur dioxide (70 ppm) and benzoic acid (120 ppm). Adding

preservatives such as Sulphur dioxide and benzoic acid can increase


68

the shelf life of RTS beverages. Colouring matter and clarifying agents

can be added to the product, to increase attractiveness and addition of

flavouring ingredients is allowed only in the product prepared by using

mango.

Therefore, this research was carried out to formulate an acceptable

quality of RTS beverage using watermelon and pineapple at different

combinations considering SLS requirements and to assess the

chemical, organoleptic and microbial qualities of pineapple blend

watermelon RTS beverage after formulation and during storage.

MATERIALS AND MEEHODS

Undamaged, disease free, healthy, mature and ripe watermelon

(Variety - Thillina) and pineapple (Variety - Mauritius) fruits were

purchased from the Commercial Horticultural Farms of the

Department of Agriculture, Sri Lanka.

Juice preparation

Fruits were washed with clean running water to remove dust particles

and to reduce the microbial load on the surface of the fruits. Peeled

pineapples were crushed with mixer cum juicer for the extraction of

juice. Watermelons were cut with the help of stainless steel knives into

pieces and seeds were removed manually. The fleshes were passed

through the juicer for extraction of juice. Six formulations were

prepared by mixing watermelon juice and pineapple juice in different

ratios with sugar, acidity as anhydrous citric acid and 70 ppm of

potassium metabisulphite (KMS). The amount of added sugar, citric


69

acid and potassium metabisulphite were kept as same for each

treatment.

The prepared beverages were filtered through the strainer (200 µm) to

get a clarified juice and filled in previously sterilized glass bottles (200

ml) leaving 2.5 cm head space and sealed airtight by crown corking

covered with aluminium foil. Then the bottled juice were sterilized at

105°C for 10 min and cooled to room temperature of 30±1°C.

The following treatment combinations were formulated:

T1 - RTS beverage with 100% watermelon juice

T2 - RTS beverage with 90% watermelon juice and 10% pineapple

juice


juice


juice


juice


juice

Chemical Qualities

Chemical qualities of the RTS beverage were analysed using

recommended standard AOAC methods (2002). The titrable acidity

was determined by titrating the RTS beverages of various juice


70

combinations with 0.1N NaOH and the results were expressed as

percentage of anhydrous citric acid.

Ascorbic acid content of beverages was titrimetrically estimated by

indophenol dye method. The pH was determined by an Electronic pH

meter (Mettler Toledo, UK). Lane-Eynon method was performed to

determine the total sugar content of the formulated beverages.

Hand-held refractometer (ATAGO-S-28Emodel) was used to estimate

the total soluble solids (TSS) and the values were expressed as °Brix.

The analyses were replicated thrice.

Microbial Test

The prepared beverages were studied for microbial quality and safety.

The total microbial load was calculated by standard plate count

method. The standard plate count was done according to the method

described by Arachchi (2003) in raw mango RTS beverages.

Sensory Evaluatio

In sensory evaluation, the samples were subjected to nine-point

hedonic scale test and the acceptability of samples was judged by 11

trained panelists to determine sensory preference. The sensory

characteristics such as colour, flavour, taste, thickness and overall

acceptability of the RTS beverages were judged by the panelists.

Shelf Life Evaluation

RTS beverages were subjected to storage studies at room temperature

for a period of 3 months by drawing samples at bimonthly intervals to


71

evaluate changes in chemical, organoleptic parameters and microbial

spoilage.

Statistical Analysis

Data obtained in chemical analysis were subjected to Analysis of

Variance (ANOVA) and mean separation was done with Duncan’s

Multiple Range Test (DMRT). Descriptive statistics was done on

sensory attributes and the means were compared using the Tukey’s test

(p<0.05).


Titrable Acidity

According to the Sri Lanka Standard Institute Specifications, the limits

of acidity for RTS preparation are 0.3-1% as anhydrous citric acid

(SLS 729:1985). The titrable acidity of the RTS beverage samples

varied significantly and increased from 0.16 to 0.42% with increase the

concentration of pineapple juice from 10 to 30% as shown in Figure 1.

This can be attributed partly to the contribution of the inherent acid

naturally present in the pineapple juice. Pineapple is acidic, which 87

% is citric acid and 13 % is malic acid. According to Samson (1986),

the fresh pineapple juice has 0.7-1.6 g citric acid /100 ml.


72

Note: The values are means of triplicates

Vertical bars indicate the standard errors.

Ascorbic Acid

Note: The values are means of triplicates

Vertical bars indicate the standard errors.

Ascorbic acid is an essential nutrient for humans because it aids in the

synthesis of collagen in addition to protecting against oxidative

damage. Watermelon juice contains 5.81 mg/100 ml of ascorbic acid

(Patil, 2000).

0

0.1

0.2

0.3

0.4

0.5

T1 T2 T3 T4 T5 T6

Tit

rab

le a

cid

ity

(as

% C

itri

c aci

d)

Treatments

Figure 1: Titrble Acidity of the Freshly Made RTS

Beverages

0369

121518

T1 T2 T3 T4 T5 T6Asc

orb

ic a

cid

(mg/1

00 m

l)

Treatments

Figure 2: Ascorbic Content of Freshly Made RTS

Beverages


73

The ascorbic acid content increased significantly from 5.64 to 15.63

mg/100 ml with an increase in the concentration of pineapple juice

from 10 to 30% in the RTS beverage formulations (Figure: 2). These

findings are reported that maximum ascorbic acid (15.9 mg/100 ml

juice) was recorded in RTS beverage with 70% watermelon juice and

30% pineapple juice.

pH

The presence of free hydrogen ions and buffering capacity of the juices

influence the pH value of the beverage (Shubhangi, 2002). The pH of

freshly made pineapple blend watermelon RTS beverages was below

5.25. According to DMRT, the pH reduced significantly (p<0.05) with

the increasing concentration of pineapple juice in RTS beverages is

shown in the Table 1. The highest pH value 5.22 was obtained in the

treatment T1. The treatment T6 (RTS beverage with 70% watermelon

juice and 30% pineapple juice) had the least mean value.

Table 1: The pH and TSS of freshly made Pineapple blend

Watermelon RTS Beverages

Note: The values are means of triplicates ± standard error

Treatments pH TSS ( °Brix)

T1 5.22±0.002a 13.1±0.02a

T2 5.10±0.023b 13.5±0.03b

T3 4.90±0.003c 13.9±0.09c

T4 4.76±0.002d 14.0±0.06d

T5 4.64±0.016e 14.2±0.04e

T6 4.61±0.016f 14.5±0.06f


74

The results generally showed that the higher acidity, lower pH of

pineapple RTS beverages. Similar study conducted by Awsi Jan and

Dorcus (2012) found that there is a corresponding reduction in pH as

the acidity increased in pineapple Juice blend with carrot and orange

juice.

Total Sugar

Sugars, acids and their interactions are important to sweetness,

sourness and overall acceptability in RTS beverages. The minimum

total sugar content (as sucrose) for RTS preparation is 5% (SLS

729:1985). Sugars, acids and their interaction are important to

sweetness, sourness and overall acceptability in RTS beverages. As

shown in Figure: 3, the total sugar content of watermelon was 5.25%.

Schmidt et al, (2005) found that the watermelon juice contained 32.2%

fructose, 9.3% glucose, and 27.1% sucrose for a total of 68.6% of

sugars on a dry matter basis.

Note: The values are means of triplicates. Vertical bars indicate the standard errors.

0

3

6

9

12

15

T1 T2 T3 T4 T5 T6

Tota

l S

ug

ar

(%)

Treatments

Figure 3: Total Sugar Content of Freshly Made RTS Beverages


75

According to DMRT, the total sugar significantly (p<0.05) differed

between each treatment. The treatment T6 had highest mean value and

the treatment T1 (RTS beverage with 100%watermelon juice) had a

least mean value at 5% level of significance.

Total Soluble Solids (TSS)

TSS:Acid ratio is often better related to palatability of the fruits than

either sugar or acid alone. The recommended TSS for commercial RTS

production is 15°Brix (SLS 729: 1985). The TSS of RTS beverage

formulation was adjusted initially. According to the DMRT, the TSS

increased significantly (p<0.05) with the increasing concentration of

pineapple juice in RTS beverages is shown in the Table 1. The highest

TSS value 14.5 (°Brix) was obtained in the treatment T6. The treatment

T1 (RTS beverage with 100% watermelon juice) had the least mean

value of 13.1 (°Brix).

Microbial Test for Freshly Made RTS Beverages

In the freshly made beverages, no bacterial growth was observed

immediately after formulation. Therefore, there was no total plate

count in these samples. Carter et al. (2007) reported that many products

that could safely be maintained sterile by pasteurization process alone

could be doubly preserved by the addition of potassium metabisulphite.

The sulphite inhibits yeasts, moulds and bacteria (Doughari and

Elmahmood, 2007). Therefore, the heat treatment was sufficient to

destroy initial microbial load in the formulated fruit drinks.


76

Sensory qualities of Freshly Made RTS Beverages

Sensory evaluation was made through panel of 11 trained judges. The

panel evaluated colour, flavour, taste, thickness and overall

acceptability. A 9-point hedonic scale was used for this purpose. As

shown in Figure: 4, the sensory evaluation of the RTS beverage

revealed that, there were significant differences between the treatments

as the concentration of pineapple juice was increased from 10 to 30%

for colour, flavour, taste, thickness and overall acceptability at 5%

level of significance according to General Linear Models (GLM).

CONCLUSION

Based on the chemical analysis of the freshly made RTS beverages,

there were significant increase in titrable acidity, ascorbic acid, total

sugar and total soluble solids (TSS) and a significant decrease in pH

with the increase in the concentration of pineapple juice from 10 to

30% in RTS beverages. According to the microbial test, the total plate

0

5

10T1

T2

T3

T4

T5

T6

Figure 4: Sensory Characteristics of Freshly Made Pineapple Blend Watermelon RTS Beverages

colour

flavour

taste


77

count was observed after three months of storage. Among the

inoculated samples, there were shown some colony forming units

(bacterial growth).

There was no drastic effect on the quality of the product due to

microbial growth in three months at ambient temperature. Therefore, it

is safe for consumption upon three months of storage. The sensory

analysis revealed that, there were significant (p˂0.05) differences for

the organoleptic characters between the treatments. According to

Tukey’s test, the highest overall acceptability was observed in the RTS

beverage with 80% watermelon juice and 20% pineapple juice. The

RTS beverage with 100% watermelon juice had low organoleptic

characters and overall acceptability than the other RTS beverages.

Therefore, the RTS beverage formulated with 80% watermelon juice

and 20% pineapple juice is the best combination for the commercial

preparation of pineapple blend watermelon RTS beverage without any

significant changes in the nutritional, chemical and sensory

characteristics with extended shelf life.

REFERENCES

Anon (2008). Watermelon. Federal Agriculture Marketing Authority.

News Letter No: 27. Tropical Research and Training Institute,

Chatham, UK.

AOAC (2002). Official Methods of Analysis. (17th Edn).Association

of Official Analytical Chemists. Washington, USA.

Arachchi, V. K. (2003). Incidence and severity of food borne

pathogens in raw mango RTS beverages. Journal of Food

Protection. 9: 2172–2198.


78

Archer, M. C. and Tannenbaum, S. C. R. (1970). Vitamins. In:

Nutritional and Safety Aspects of food Processing. Marcel

Dekker, Inc., New York, UAS. pp. 53-60.

Awsi, J. K. and Dorcus, M. R. (2012). Development and Quality

Evaluation of Pineapple Juice Blend with Carrot and Orange

juice. International Journal of Scientific and Research

Publications. 2(8): 2250-3153.

Carter, H. W., Charley, V. L. S. and Bristol. C. (2007). The

Preservation of fruit Juice Products with special reference to

nutritional value. Journal of Food Safety. 88: 112-118.

Cyril, P. U., Sandrine, T. and Schini, L. H. (2009). Evaluation of

different fruit juices and optimization of a red watermelon fruit

juice blend. Food Functions. 21:245-251.

De Carvalho, J. M., G. A. Maia, R. W. and De Figueredo (2007).

Development of a blended non-alcoholic beverage composed

of coconut water and cashew apple juice containing caffeine.

J. Food Qual. 30: 664-681.

Doughari, J. H. and Elmahmood, A. M. (2007). Effect of some

chemical Preservatives on the shelf life of Sobo drink. African

Journal of Microbiology. 2:5-6.

Patil, D. (2000). Development of value added nutritious products from

watermelon. M.Sc. Thesis. University of Agricultural

Sciences, Bangalore, India.

Samson, R. B. (1986). Changes in nutritional composition and sensory

qualities of the fresh pineapple juice during heat treatment and

storage. Journal of Food Quality. 29: 165-173.

Shubhangini, A. J. (2002). Nutrition and dietetics. Tata McGraw Hill

Publish Company Ltd, India. pp. 488-58. Smooth Cayenne

cultivars. Food Chemistry 53: 75-79.

Sri Lanka Standard Institute (1985). Standards and Specifications for

ready-to-serve fruit drinks. SLS 729:1985, Colombo.

http://xlink.rsc.org/?DOI=c1fo10040h




Economic Research

Volume 3 (1)

June 2015: 79-111

Sri Lanka Forum of


SLJER

Asiri D. Vitharana

________________________________________________

79

THE SOLID WASTE MANAGEMENT FOR

SUSTAINABLE DEVELOPMENT: A CASE

STUDY OF HAMBANTOTA MUNICIPLE

COUNCIL AREA IN SRI LANKA

Abstract

The study explores the importance of Solid Waste Management (SWM) for

Sound Environmental Development (SED) in Hambantota Municipal Council

(HMC) area in Sri Lanka. Data for the study was collected using integrated

research tools including Semi-structured interviews, self-completion

questionnaires, focus group discussions and non-participatory observations.

The ‘Random Sampling Method’ was used to select households and business

population while ‘Purposive Sampling’ was used to select respondents for

the interviews and focus groups. The study revealed that lack of awareness,

little synergy between local authorities and the community and challenges of

limited resources, facilities and funds and illegal dumping for proper SWM

process were the dominant problems in SWM. In this regard, the paper

suggests an integrated SWM process, with awareness programmes and

cooperation between authorities and community for the process.

Implementing SWM in to Development policies for Sustainable Development,

has been further suggested.

Keywords: Sustainable Development, Solid Waste Management, Hambantota

Municipal Council, Sri Lanka

Asiri D. Vitharana University of Ruhuna, Matara, Sri Lanka

[email protected]


80

INTRODUCTION

Over the time, countries experience the development with growth of

industrialization, trade and internationalization. To some extent, a

discussion has emerged whether this development is sustainable.

Recently, most development practitioners have focused on

‘Sustainable Development’ (SD)1 as an environmental concept placing

the emphasis on intergenerational equality (Carter, 2001). According

to World Conservation Strategy (WCS), development depends on

environmental conservation (Adams, 2009). However, the human

way of life has put an immense pressure on the environment, because

of high consumption patterns accelerating in parallel with the

economic development.

Solid waste (SW) has become a major negative consequence of

development in recent decades. SW largely affects the natural

environment, and poses a serious threat in achieving Sound

Environmental Development (SED) or so-called Sustainable

Development (SD) (Anand, 2010). Therefore, sound SW management

is one of the necessities in development process.

Sri Lanka may provide an example of this issue, with new

development processes under the government's major development

plans, since the environment is challenged with rapidly generating SW.

Consumer behaviour has generated a significant amount of SW in

1 As guided by the United Nations (UN) Brundtland Commission, sustainability is ‘a

system which can be considered sustainable if it ensures that today’s economic

development is not at the expense of tomorrow’s development perspectives’ (Malik

et al, 2011).


81

recent years. According to the National Environmental Action Plan

(NEAP), management of solid waste has been identified as one of the

major challenges for sound environmental development in Sri Lanka.

However, when considering the sound environmental

development/Sustainable Development in Sri Lanka, it is important to

assess the impact of development project on the environment,

especially for the generation of Solid Waste, as it is the first and the

most significant outcome of the development. Hambantota

development projects were implemented during last few years. There

are no proper research studies on the impact of Solid Waste

Management for sound environmental development in the area. Even

though the development projects have mentioned about sustainable

development, there is no proper description and procedure or

discussion on the importance of Solid Waste Management for

sustainable development process.

As Hambantota district is being developed under a master

development plan, it is important to direct this development towards

SED/SD for future generations. Therefore, the research considers

SWM practices and its impact on SED/SD in Hambantota Municipal

council (HMC). The study further focuses on identifying the

importance of SWM for the formulation of on-going development

processes towards sustainability. The objective of this study is to

explore the existing SWM practices, analysis their impact on

sustainable development in the study area finally make

recommendation to improve the SWM processes.


82

CONCEPTUAL FRAMEWORK

Figure 1 depicts a Conceptual Framework (CF) to explain the link

among SWM, Environmental Conservation and Sustainable

Development. The framework elaborates that development occurs with

growing economy and increasing population generates solid waste,

which in turn causes environmental pollution and degradation.

Figure 1: Conceptual Framework

Source: Created by Author

Waste separation, recycling and composting can be used as methods

for SWM with integrating of such measures as legal structure,


83

awareness raising, attitude change, and cooperation between

authorities and community to mitigate the negative impacts on the

environment. Integrated, sustainable SWM provides the ground for

environmental conservation and in turn socio-economic development.

Efficient waste management will facilitate steady state growth and to

achieve the sustainable development/Sound Environmental

Development at the end.

LITERATURE REVIEW

Over the past 30 years, most of countries in the world have achieved

rapid economic development in term of improving the well-being of

people and raising their living standards. Despite some remarkable

progress, there are also pressing constraints on development with

entrenched negative trends means that sustainability is remaining

question (Dalal-Clayton, and Bass 2002). Development practitioners

now recognize Sustainable Development as a matter of environmental

conservation that should be followed by social and economic

development. Environmental sustainability is a necessity for SD in

order to balance the three pillars of economic and social development

with environmental protection (Malik et al, 2011). The Recognition led

to initiate discourser and create the United Nations Environmental

Programme (UNEP) and United Nations Environment Department

(UNED). Since then, worldwide acceptance of the importance of the

environmental issues has grown enormously (Dalal-Clayton and Bass

2002). Concurrently, the report of the World Commission on

Environment and Development (WCED - the Brundtland commission)


84

was introduced in response to increasingly informed analysis of the

link between environment and development (Dalal-Clayton and Bass

2002).

Focusing on environmental aspects of the Sustainable Development,

the crucial condition appears to be that of reducing environmental

degradation and resource consumption combined with economic

development (Clini et al., 2010). However, accelerating environmental

degradation indicates that the world is facing a stronger sustainability

challenge than ever (Ayres, 2007). The achievement of sustainability

in national development requires a strategic approach, which is both

long term in its perspective and integrated or joined up in linking

various development processes; so that they are as sophisticated as the

challenges are complex (Dalal-Clayton and Bass, 2002). The

important message is that environment and the socio-economic

development are so intricately linked that development cannot be

sustained without considering its environmental dimensions (UNEP,

2013). Therefore, to achieve Sustainable Development goals,

countries should develop attentiveness towards sound environmental

development and mitigate impacts on the environment.

Sustainable Development rejects the practices that support depletion of

natural resources and that leaves future generations with poorer

prospects (Repetto, 1986). Human actions have caused many

challenges to the environment, especially with tremendous amount of

waste generated due to population growth, industrialization and

increasing consumption patterns. Unmanaged development influences


85

the production of wastes in different ways and strongly affects the

environment all over the world. This cannot be avoided as long as

waste generation is a necessary reality associated with economic

development (Vallero and Letder, 2011). Waste is described as

something that has ‘no value,’ something which is ‘useless’ and

something that the owner wants to ‘discard.’ The words Rubbish,

Garbage, Trash or refuse are often used as synonyms for waste. The

UNEP (2009) classifies waste as; Municipal Solid Waste (MSW)

including Plastic waste, Construction and Demolition waste (C&DW),

Hazardous solid wastes (HW), Bio-Medical Waste (BMW) and

Electronic waste (E-waste). Out of all these wastes, many countries are

facing serious problems with the increasing amount of Solid Waste

(SW) SW is viscid, dense, hard and substantial materials. SW has been

further called ‘Municipal Solid Waste’ (MSW). The urban

development process, population growth and consumption patterns

have increased the generation of SW/MSW in many countries. The

institutions both Government and non-government have responsibility

to manage the waste in order to mitigate the possible impact on health

and environment of the people.

According to the Schübeler, (1996) Solid Waste Management (SWM)

is an essential task for ensuring public health and well-being, quality

and sustainability of the urban environment and efficiency and

productivity of the urban economy. As Chandrappa & Das (2012)

described, SWM processes may vary due to different lifestyles of the

inhabitants and type of the natural resources in the respected region.


86

Therefore, the strategy towards SWM should be differentiated

accordingly.

The aim and purpose of Solid Waste Management is to ensure a sound

environment and increase resilience. The underlying principle of the

Basel Convention (1989), points out that the environmentally sound

management is an approach that has the potential to bridge the makers

to the breakers and thus contributing worldwide to the protection of

human health and the environment. Therefore, Sound Environmental

Waste Management should focus on integrating all practical steps to

protect human health and the environment from the adverse effects of

all kinds of wastes (Portas, 2003). The Integrated Solid Waste

Management 2 is a new, broad, international consensus on the

Management of SW/MSW in sustainable ways and the concept was

introduced by the organization WASTE first and was later developed

and expanded by the

Collaborative Working Group on Solid Waste Management in Low

and Middle-Income Countries (UN Habitat, 2010). ‘Integrated Waste

Management’ is a framework of reference for designing and

implementing new waste management systems and for analysing and

optimizing existing systems (UNEP, 2005).

2 ‘Integrated Waste Management’ is a framework of reference for designing and

implementing new waste management systems and for analysing and optimizing

existing systems (UNEP, 2005).


87

According to McDougall et al (2001), ISWM uses a range of collection

and treatment methods. All materials in the waste stream are handled

in environmentally effective, economically affordable and socially

acceptable ways. The Integrated Approach to Solid Waste

Management can deliver both environmental and socio-economic

sustainability. Countries that can achieve SW/MSW with ISWM may

move towards sustainability.

A healthy environment has the ability to serve as a sink or waste

repository. It can absorb, redistribute and transform waste, produced

by human action or natural phenomenon, into harmless forms (Coward,

1995; Park, 2001; Hawken, 2003). However, the process of unbalanced

development has its price in the form of an increased pressure on

natural resources through inefficient and wasteful utilization (Anand,

2010). As such, the amount of waste in recent decades has exceeded,

“the environment’s ability to absorb the waste” (Ritzre, 2010).

The significant issue with this increasing amount of waste is that they

have depleted the ecosystems by pollution and degradation. The

capacity of many ecosystems to provide services has been reduced and

it has had profound negative impacts on opportunities for Sustainable

Development around the planet (UNEP, 2007). In this situation, the

future options for sustainable development become limited or eroded

as the ecosystem products and services are not managed effectively and

efficiently (UNEP , 2007).

Environmental conservation, protecting the ecosystem and services

should be the main concerns. Especially, as discussed above,


88

SW/MSW may have significant impacts on the environment, and

managing them properly is a necessary prerequisite for making

development more sustainable. The well-formed Solid Waste

Management is the only way to reduce the impact of SW/MSW on the

environment. Due to the implementation of successful Solid Waste

Management, both public health and the quality of the environment

will benefit directly and substantially (UNEP, 2005)

The escalating development process in Sri Lanka has significant

impacts on the environment. Today the country’s bio-diversity is

endangered and environmental degradation has increased due to

unmanaged development and economic growth. Population growths,

consumption of ecosystem products and services have caused

significant impacts on the environment, including the generation of

large amounts of SW/MSW.

SW/MSW is a growing issue in urban development in Sri Lanka and is

aggravated due to the absence of proper solid waste management

systems at local authorities (CEA, 2005). According to the Sri Lanka

National Report to the WSSD (2002), there are a number of critical

environmental problems and issues caused by SW/MSW such as:

deforestation, coastal erosion, soil erosion, water pollution and

atmospheric pollution due to development processes. Therefore,

SW/MSW is considered as one of the main environmental problems

which affects urban air pollution, water pollution and soil erosion in

Sri Lanka (Wanasundara, 2002).


89

The Ministry of Environment & Natural Resources of Sri Lanka

launched a national level Solid Waste Management programme called

‘Pilisaru’ (renovation). The Project was undertaken to find solutions

to the increasing solid waste problem with the participation of other

organizations such as the Urban Development Authority, private

institutions, NGOs & experts in the field in Sri Lanka. Under this

Project, waste management will be managed by adhering to reduction

of waste generation by reuse, recycling and resource recovery to the

maximum extent possible followed by appropriate treatment and

finally the disposal of residual waste in an environmentally sound

manner (CEA, 2008).

METHODOLOGY

Study Area

The study area of this study is Hambantota city which is governed by

Hambantota Municipal Council (HMC) since 2011. The city located in

Hambantota District of Sothern dry-zone of Sri Lanka. It includes 7

Grama Niadhari Divisions (GNDs): Koholankala, Keliyapura,

Siribopura, Samodagama, Hambantota East, and Hambantota West.

The total population in HMC is 23,090 with 5852 households and 565

registered Business Places.

Hambantota has rich ecosystems including a diverse flora and fauna

and various environmental resources as shown in Map 1. The

conservation of these ecosystems is a concern of many environmental


90

practitioners and policy makers in order to support sustainable

development in the country. However, as Hambantota area is being

developed under the government greater development plan (See maps

2), there are potential impacts on the environment. In this situation, the

study focuses on the impact of SW/MW on the environment in order

to discuss the importance of proper SWM process for SD/SED in the

study area.

Map 1: ecosystem in Hambantota

l

Source: Urban development authority


91

Map 2: Greater Development Plan in Hambantota

Source: Urban Development Authority (2011).

Data and Analysis

Both qualitative and quantitative methods including the semi-

structured interview, self-completion questionnaire, focus group

discussions and non-participatory observation were used to collect

data. Government reports, municipal council reports and many other

documents from different organizations and academic papers regarding

waste management and its impact on the environment were used to

collect secondary data. Random sampling was used to select 69

households and 59 business populations for SCQ, of 5852 households

and 412 business places with expected errors 0.12%. The sample size

was determined by using a mathematic function ( n =N

1+ N (α)^2 (See

table 1). Purposive sampling was used to select participants for

interviews and focus groups and self-observation areas. Qualitative


92

data was analysed with the Thematic Analysis Approach, where the

researcher looked for patterns to build up themes.

Table 1: Sample Size

GND Number of HHs/BPs Equation Sample

Total

Households 5852 𝒏 =𝟓𝟖𝟓𝟐

𝟏+ 𝟓𝟖𝟓𝟐(𝟎.𝟏𝟐)^𝟐, = 68.63 69

Business

Places 412 𝒏 =

𝟒𝟏𝟐

𝟏+ 𝟒𝟏𝟐(𝟎.𝟏𝟐)^𝟐, = 59.42 59

Koholankala

Households 756 𝑛 =68.63

5852× 756 = 8.86 9

Business

Places 24 𝑛 =

𝟓𝟗.𝟒𝟐

412× 33 = 4.75 5

Siribopura


5852× 1675 = 19.64 20

Business

Places 56 𝑛 =

𝟓𝟗.𝟒𝟐

412× 56 = 8.07 8

Keliyapura


5852× 503 = 5.89 6

Business

Places 14 𝑛 =

𝟓𝟗.𝟒𝟐

412× 14 = 2.01 2

Samodagama


5852× 432 = 5.06 5

Business

Places 1 𝑛 =

𝟓𝟗.𝟒𝟐

412× 1 = 0.14 0

Hambantota

East


5852× 1757 = 2.59 2

Business

Places 62 𝑛 =

𝟓𝟗.𝟒𝟐

412× 62 = 8.94 9

Hambantota

West


5852× 221 = 20.60 21

Business

Places 212 𝑛 =

𝟓𝟗.𝟒𝟐

412× 212 = 30.57 30

Mirijjawila


5852× 508 = 5.95 6

Business

Places 34 𝑛 =

𝟓𝟗.𝟒𝟐

412× 34 = 4.90 5

Source3: Author Calculation

3 Author used statistical data from Sri Lanka CSD and HMC register


93

When the final classification of the themes and discussion of the

findings was done with regard to the literature review and the data from

documents reviewed. The research findings supported by the

qualitative data, which was used to find relationships such as

deviations, correlations, regressions and tendencies of given statistics.


Waste Management Practices in HMC

The sources of SW include residential and commercial activities in

HMC. In this study, waste has been categorized into nine categories

including paper and carton, plastic, food waste, tins and cans, fiber

bags, glass, tree leaves, coconut shells and charks & ashes.

Figure 1: Characteristic of Waste Disposal

Source: Survey Data


94

As shown in figure 1, households in the sample mostly discard food

waste (22%), tree leaves (21%) and plastics (16%) while business

places mostly discard paper and carton (39%); plastic (29%) and food

waste (20%). Sample as a whole however, 59% from households (HH)

and 61% from business places (BP) discard organic waste and

therefore it is possible to manage these wastes with composting

processes. There are 21% of HH and 29% of BP discard plastic waste,

it is important to manage these wastes effectively as plastic materials

significantly impacts to the environment.

Illegal disposal practices have become a challenging issue in the study

area. As shown in Figure 2, 81% of HH and 43% of BP dispose (or

dump) their SW in to unauthorized places nearby water bodies,

roadsides, open spaces and use other unauthorized.

Figure.2: The Places of Waste Disposal

Source: Survey Data


95

It is found from the Focus Group Discussions that lack the supply of

public waste bins by HMC has encouraged the community to dispose

their waste in the unauthorized places. The survey further revealed that

about 96% of HH and 83% of BP do not have access to public waste

bins to dispose SW.

The observation was able to discover many illegal dumping sites in the

study area. Especially the unauthorized dumping sites near ‘Bundala’

National Park, sub roads, near water resources are challenging. The

researcher further observed that most of the unauthorized waste

dumping sites has the wind-blown polythene which caused a myriad of

nuisances to residents, to the public and to the environment.

The illegal dumping has further increased due to the lack of proper

waste collection process by HMC. The HMC has not broadened their

waste collection process, since they became a Municipality with seven

GNDs. Only 19% of the HH and 49% of BP has access to Municipal

Council waste collection tractor for disposal. This indicates that HMC

is ineffective and inefficient in providing waste collection facilities to

the community.

During the survey period, three important components which are

affecting towards both the primary disposal practices, waste collection

and transportation in the study area. These include: a) lack of funds,

facilities and recourses b) lack of education information and awareness

and c) lack of synergy between the local authorities and civil society.

Respondents in the focus group discussion suggest that raising funds,

improving infrastructure and increasing equipment and human


96

resources would be a positive impact on the delivery of the SWM

services. In contrast, HMC members and responsible authorities for

SWM view that these suggestions will impose an economic burden to

the HMC and emphasized the needy of Central Government support to

be a solution for the lack of financial resources.

Moreover, HMC has the responsibility to use the allocated funds

effectively and efficiently to improve facilities and to increase

resources and human resources for SWM processes. This can reduce

the illegal dumping practices by provisioning adequate skips and

public waste bins and improving waste collection processes.

Disposal is the final element in the solid waste management process.

This is done either through land filling or land spreading. Sanitary land

filling sites are the most appropriate place of final disposal for solid

waste. But in the case of HMC, there is no sanitary landfill site for final

disposal. In the focus group discussion officer in charge of the site

pointed out that, “We do not identified sanitary landfill site and

therefore we use the open dumping strategy. We don’t have facilities,

resources or funds to maintain such a sanitary land fill for the moment.

We just dump all the waste in to the land we have. We do separate

recyclable waste to some extent and for the lefts we use open dumping”

The workers in Hambantota Integrated Solid Waste Management

Centre (HISWMC) is collecting some selected wastes using tractors

for composting processes before the final dumping. As pointed out by

workers in the centre, they are used to recycle plastic waste. Thus, the

observation revealed that the process is not effective or efficient at the


97

date. However, according to information gathered by SSI revealed that

impossibility to separate waste on the site or compost due to the lack

of facilities, equipment and labours as well as inability the HMC to

maintain a sanitary land fill. The photograph 1 below shows the

landfill site of HMC. Moreover, the composting facilities in HMC have

been ineffective and insufficient, due to inadequate monitoring of the

composting process and quality of the compost being produced.

Photo 1: Landfill Site of HISWMC

Source: Author (Self Observation)

These evidences call for necessity of adopting the Integrated Solid

Waste Management strategy to ensure effective Solid Waste

Management in HMC. The landfill site should be properly managed to

avoid heaping of waste and burning. Hambantota Integrated Solid

Waste Management Centre (HISWMC) should be adequately

resourced by the government to ensure efficient and effective waste

management in the area because of the massive government led

development project has been initiated in the area. The HMC should

coordinate with corporate bodies like National Solid Waste

Management Centre as well as with NGOs to obtain the technical


98

supports and financial resources to sustain the institutions. Improving

technical knowledge about the SWM processes and developing a

monitoring process are important for Sustainable Waste Management

process in HMC.

Waste Separation, Recycling and Reuse

Waste separation practices in the HMC area are similarly low. Results

reveal that community members do not separate waste as the waste

collectors again mix them together with the transportation. Yet,

willingness of community members for waste separation is higher, if

there is a programme with an ISWM strategy. 81% of households and

63% of business places are willing to separate waste while 17% of

households and 15% of business places were not. Research further

reveals that some of the householders recycle their waste and sell them

to local waste collectors in order to get an income. However, lack of

opportunities and places to sell these collected waste decreases the

interest of recycling industry.

The findings suggest that the importance of improving the recycling

processes with Integrated Solid Waste Management (ISWM) process

of HMC. Development of recycling processes in the study area can be

done through organizing community awareness programmes such as a)

Providing communities with information about how to sort waste b)

economic and social benefits can gained form recycling process, and

how they can participate in the designing of the programs. The success


99

of recycling not only depends on participation levels of the local

community, but also on the efficiency of the equipment and

infrastructure for the SWM process. Finally, the synergy between

community and local authorities should improve to develop

cooperative processes. Citizens should share responsibility with the

municipality for decision making on the SWM in the study area.

The Evaluation of Environmental Impacts of Solid Waste

Management for SED/SD in Hambantota Municipal Council

According to Wanasundara (2002), the main environmental problems

that Sri Lanka faces include urban air pollution, water pollution,

erosion and indiscriminate garbage disposal. There can be found many

places where illegal dumping had been done near to the

environmentally sensitive areas in Hambantota. Due to these wastes,

significant impacts on soil and land could be identified in the HMC

area.


100

Photo 2: Waste on the Water Bodies

Photo 3: Burning Waste Causes Air Pollution

Moreover, there are impacts of SW/MSW on water bodies. Water is a

basic requirement for sustaining the life. Yet, due to the pollution

caused by waste, concentration of dissolved carbons, heavy metals,

biohazards such as bacteria and virus and other nutrients will increase

in the water sources resulting in loss of bio-diversity and making the

water unsafe for consumption.

As illustrated in the photograph 2, the channel system and canals are

filled with waste. Furthermore, the research identified possible


101

impacts of SW/MSW on the air, due to landfilling and waste burning

(see photo 3).

As discussed above, most of the waste dumping sites in the study area

were illegal and has accelerated environmental pollution with high

level of waste on water bodies and land. It is further found that a

significant correlation is among the variables: "the wastes dumping

place" and "waste on land" (soil pollution), "waste on water bodies"

(water pollution) and "burning wastes" (air pollution). Dumping waste

on illegal lands has increased the amount of ‘waste on land’ and

‘burning waste.’ Waste on land has also increased the amount of waste

on the water bodies.

Table 2: Correlation between Wastes on Emptied Place and Waste

on Land, Water Bodies & Burning Waste

Source: Survey Data

The table 2 shows the correlation coefficient (r) between waste on the

land and the waste dumping places is 0.481 and it is significant at 0.01


102

Level, indicates a strong relationship between two variables. Higher

correlation between the waste on water bodies and emptying place of

solid waste was also observed [(0.310 (0.01)]. The value of correlation

coefficient between the burning waste and the waste emptied place is

0.297 which is significant at the 0.05 confidence level. The table 2

further reveals a positive relationship with waste on the land and waste

in the water. This means that the waste on the land increases the waste

in the water eventually cause water pollution.

Furthermore, the final disposal site also has significantly impacted on

the land and air pollution. As expressed by the community participants

the ordo and dark water has made nuisance in the final disposal site.

The Mosquitoes, Cockroaches and Rats have caused lots of health

problems including the dengue and fever. The improper disposal of

polythene causes deaths of animals such as birds, dogs, and some other

wild animals.

In depth inquiries about this issue was done with members of

participating to Focus Group Discussions (FGDs). The discussion

reveals that even though people are aware of these health issues, lack

the knowledge about bad impact of improper waste management

practices and related impacts on the environment and public health.

This finding (e.g. there is not enough information about environmental

impacts of solid waste management) can be confirmed by the data

collected from the sample survey. It is evidence from the survey data

that only 7% of households and 17% business places agreed about the

idea that there is adequate enough information while 93% of


103

households & 83% of business places said that there is no enough

information about SWM impact on the environment. Participants to the

face to face Interviews also pointed out that lack of any system to

obtain the information to the community in the HMC.

With the findings, the study reveals the negative impacts of SWM

processes on the local environment. Implementation of proper SWM

process is the only way to manage SW/MSW. Therefore, the

responsible authorities should take steps to reduce environmental

impacts of SW, by implementing an Integrated Solid Waste

Management strategy with the support of the central government and

the central environmental authority.

As lack of knowledge about SW, SWM, and SWM impact on

environment and health were significant, awareness campaigns,

educational workshops and waste recycling and composting trainings

should be conducted for community members. Moreover, in order to

decrease the usage of natural resources and waste disposal, the waste

prevention strategies should be promoted with an ISWM strategy.

The developmental process has accelerated the generation of

SW/MSW and has increased impacts on the environment and

sustainable development and vice versa. 56% of the households and

47% of business places agreed that the new development process has

impacts on solid waste generation in the area while 28% households

and 12% business places have no idea about it. 16% of households and

41% of business places did not agree.

http://dictionary.reference.com/browse/vice+versa


104

Since the SW/MSW has significant impacts, as discussed in the

previous sections, the generation of waste with the growth of

development should be a concern if it is a goal to achieve sustainable

development. Further discussion with the respondents of the interviews

and focus group discussions revealed that there are environmental

challenges due to new development processes in the study area. As

expressed by a community leader, “there was no much impact on the

environment before. But, with new development project, impact of SW

was significant” (Field work, 2014).

Focusing on these environmental aspects of the sustainable

development, the crucial condition appears to be that of reducing

environmental degradation caused by SW. Thus, the new

developmental process in the study area has significant negative

impacts on the environment with rapidly generated solid waste. Chart

3 shows the cross tabulation of the two variables; the impact of the new

development process on SW generation and the waste on the land and

water. Moreover, a Pearson Chi-square test was conducted to examine

the statistical significance of the cross tabulation table and find out

whether there is a relationship between the impact of the new

development process on SW generation and the waste on the land and

water. As shown in table 2, the results reveals that there is a significant

relationship between the two variables.


105

Chart 3: The Impact of the New Development Process on SW

Generation and the Waste on the Land and Water

Source: Research Statistic


106

Table 3: New Development Impact on SW Generation, the Waste

on the Land & Water

Source: Research Statistics

Research findings further reveal that there is a gap in environmental

coverage in the waste management law. The community is not well

aware about the existing laws and regulations. The need for regulation

and policies regarding SWM is a prevalent concern of community

leaders and the council members. In this regard, the development

practitioners and policy makers should focus on improving

development plans and implementing Sustainable Development

Policies to mitigate the negative impacts of development processes on

sound environment and Sustainable Development. Reducing the gap


107

in environmental coverage of waste management laws and adopting

new Laws and regulations to improve effective and efficient SWM

process is suggested.

Community participation in development processes in the region is

essential for achieving sustainability for future generations. Increasing

communities’ knowledge and education about Sustainable

Development, Sound Environmental Development and possible

impacts of environmental pollution and degradation on SD/SED is

important. However, the important issue here is that the community

members are not aware of ‘Sustainable Development’ or ‘the Impact

of Solid Waste Management on Sustainable Development.’ Most of the

people in the study area including development agents and government

officials are also not aware of sustainable development concepts. 75%

of households and 80% of business places have not heard about

sustainable development. Only 25% of households and 20% of

business places are aware of SD.

Awareness campaigning, educational programmes and sharing

information about sound environmental SWM processes and

sustainable development are suggested as information and education

may encourage the community to act more properly.

Moreover, the development authorities should have more interaction

with the local governments as the development accelerates the SW

generation. The development authorities have the responsibility to

interact with the responsible persons for SWM and develop strategies

for better SWM processes with the development policy. Finally, it is


108

important that the responsible persons for SWM in HMC should

become involved in the process effectively and efficiently. Especially

the political bodies in HMC and managerial persons should be

involved in the activities more actively in order to carry out a better

SWM in the study area.

CONCLUSION AND RECOMMENDATIONS

The study has investigated the central importance of SWM for

Sustainable Development in Hambantota Municipal Council. The

result supports the idea that lack of proper SWM process has

significant impact on environment and Sound Environmental

Development. One of the major findings to emerge from this study is

that lack of proper SWM process which has increased illegal waste

dumping and open dumping. Research also identifies lack of education,

information and awareness of citizens and municipal leaders on the

impacts of waste on environment and SED. Lack of synergy between

members of the local authorities and civil society (local community,

Community organizations, NGOs and other organizations) was also

significant in Hambantota context.

Although the current study is based on a small sample of participants,

the findings suggest that SWM practices must be improved by using

the Integrated Solid Waste Management process with waste separation

from the domestic level, more efficient waste collection systems and

sustainable recovery and disposal practices. Considering the nature and


109

components of generated waste; the reuse, recycling and composting

processes would be more suitable to address the waste management

challenge in Hambantota. Public education, awareness and cooperation

between authorities and community should also be improved.

Since the development processes have accelerated the generation of

SW in the Hambantota area, it is necessary to implicate SWM policies

into development projects, and implement them before, during and

after the projects towards sustainable development and/or Sound

Environmental Development. Generally, the study contributes to

knowledge by determining the importance of SWM for SED/SD.

Therefore, future research should be focused on the investigation of the

importance of proper SWM process and its impact on sustainable

development process.

REFERENCES

Adams, W. M., (2009). Green Development: Environment and

Sustainability in a Developing World, 3rd Ed, New York:

Routledge Publication

Ayres, R., (2007). On the practical limits to substitution. Ecological

Economics, vol. 61, No. 1.

Carter. Neil. (2001). The Politics of the Environment: Ideas, Activism,

Policy, Cambridge: Cambridge University Press

CEA (2005).Technical guidelines on solid waste management in Sri

Lanka.

http://www.cea.lk/pdf/Solod%20Waste%20Management

%20Guidlines.pdf [Accessed on 2013/10/20]

Chandrappa, R. and Das, D.B., (2012). Solid wastes management

principles and practice, New York: Springer Publication

http://www.cea.lk/pdf/Solod%20Waste%20Management%20Guidlines.pdf

http://www.cea.lk/pdf/Solod%20Waste%20Management%20Guidlines.pdf


110

Clini et al., (2010).Sustainable development and environmental

management: experiences and case studies, New York:

Springer publication

Coward, A.G., (1995). Population consumption and the environment:

Religious and secular responses. New York: Suny press

Dalal-Clayton, D.D.B., Bass, S., (2002). Sustainable development

strategies: A resource book. London, Earthscan Publication

Letder, T.M., Vallero, D.A., (2011). Waste; A handbook for

management. Amsterdam: Academic press

Malik, Abdul et al (2011).Environmental protection strategies: an

overview. In Malik, Abdul and Grohmann, Elisabeth (Ed),

environmental protection strategies for sustainable

development New York: Springer Publication

McDougall, et al., (2001). Integrated solid waste management: a life

cycle inventory, Oxford: Blackwell

Park,C. Chris., (2001). The environment: principals and applications.

Psychology press

Portas .P. (2003).From makers to breakers: a new dimension in

worldwide waste management. In, Dhir, K.R. et al, ed.

Sustainable waste management, Thomas Telford

publication

Repetto, Robert C., (1986).World enough and Time: Successful

Strategies for Resource Management. New Haven: Yale

University Press

Ritzre.G. Ryan, J.M., (2010).The concise encyclopedia of sociology.

New Jersey : Jhon Wiley and sons

Schübeler, P., (1996). Conceptual Framework for Municipal Solid

Waste Management in Low-Income Countries, World

Bank.http://www.worldbank.org/urban/solid_wm/erm/CW

G%20folder/conceptualframework.pdf [accessed on

2013/09/25]

http://cmc.opac.marmot.org/Author/Home?author=Repetto%2C%20Robert%20C.

http://en.wikipedia.org/wiki/Hoboken,_New_Jersey

http://www.worldbank.org/urban/solid_wm/erm/CWG%20folder/conceptualframework.pdf

http://www.worldbank.org/urban/solid_wm/erm/CWG%20folder/conceptualframework.pdf


111

UN Habitat (2010). Solid waste management on the world cities: water

and sanitation on the world cities. London, Earthscan

publication

UNEP (2005). Solid Waste Management,

http://www.unep.org/ietc/Portals/136/SWM-Vol1-Part1-

Chapters1to3.pdf [Accessed 2013/03/12]

UNEP (2009), Developing Integrated Solid Waste Management Plan:

Training Manual, VO 4: ISWM Plan

UNEP (2013). Embedding the Environment in Sustainable

Development, Goalshttp://www.unep.org/pdf/embedding-

environments-in-SDGs-v2.pdf [Accessed on 2013/11/18]

Wanasundara, N.P., (2002). Culture of the world: Sri Lanka. : New

York: Marshall Cavendish

Wickramesinghe, R.H., (1990). Pollution in Sri Lanka .NARESA:

Colombo

http://dl.nsf.ac.lk/bitstream/1/5584/1/VIDU%2012_1_1.pd

f [Accessed on 2012/02/30]

http://www.unep.org/ietc/Portals/136/SWM-Vol1-Part1-

http://www.unep.org/pdf/embedding-environments-in-SDGs-v2.pdf

http://www.unep.org/pdf/embedding-environments-in-SDGs-v2.pdf

http://dl.nsf.ac.lk/bitstream/1/5584/1/VIDU%2012_1_1.pdf

http://dl.nsf.ac.lk/bitstream/1/5584/1/VIDU%2012_1_1.pdf


112


Economic Research

Volume 3 (1)

June 2015: 113-153

Sri Lanka Forum of


SLJER

D.I. J. Samaranayake

O.G. Dayarathna-Banda

----------------------------------------------------------------------------------------------------------------

113

UNEMPLOYMENT, OFFICIAL ECONOMY

AND THE DIMENSION OF THE SHADOW

ECONOMY: AN EMPIRICAL ANALYSIS

FOR SRI LANKA USING SEM APPROACH

Abstract

This study attempts to estimate the share of Sri Lankan shadow economic

activities as a percentage of official estimates, while revolving the wheel of

non-clarified zones and market functions through overstepping the

traditional official estimates. The methodology involves the estimation of

structural models to analyse a set of causes of the shadow economy and its

influence upon a series of indicators. The study introduces three Multiple

Indicator Multiple Cause (MIMIC) models namely MIMIC 5-1-2a, MIMIC 4-

1-2b and MIMIC 3-1-3a. The benchmark calculations for each model derives

a series of average values for the Sri Lankan shadow economy (SE) in the

period from 1990 to2012. Estimated data in model MIMIC 5-1-2a suggests

to evidence that the average size of shadow economy in the country is ranging

between 91% and 32% in the period from 1990 to 2012 with a decreasing

trend. By contrast, calculations for MIMIC 4-1-2b and MIMIC 3-1-3a

demonstrate a size of 14% and 52% with an increasing trend respectively.

In-depth analysis further reveals the facts that effect the share of tax on goods

and services to the government revenue and the level of public employments

tend to undermine the increasing pattern of shadow economy. Since the

unemployment rate and private employment is playing a charismatic role in

the economy, shadow economy tends to increase. Eradicating the workplace

enforcement crisis and underemployment issues may hinder the increasing

pattern. The results from re-examination of Okun’s law supports for the idea

that, less interdependence of the growth of shadow economy and official

economy and a parallel growth with shifting stages in market functions.


114

Keywords: Shadow Economy; Unemployment Rate; MIMIC Models; Benchmark

Calculations; Okun’s Law; Official Economy

JEL Classification: E26, E27, C36

------------------------------------------------------------------------------------

D.I. J. Samaranayake

Assistant Lecturer, Department of Economics and Statistics, University of

Peradeniya, Sri Lanka

E-mail: [email protected], T.P: (+94) 075-6769290

O.G. Dayarathna-Banda

Senior Lecturer, Department of Economics and Statistics, University of Peradeniya,

Sri Lanka

E-mail: [email protected] T.P: (+94) 077-9890035



Unemployment, Official Economy and the Dimension of the Shadow Economy

115

INTRODUCTION

Existence of a considerable share of shadow economic activities is one

of the major obstacles faced by the emerging nations, like Sri Lanka. It

may hinder the growth of a strong public sector as well as creation of

a functioning market economy1. Existence of Shadow Economy (SE)

imposes the pessimistic costs such as tax evasion, market distortion,

unfair competition and inefficient resource allocation in the economy.

On the other hand, it provides an important safety net for individuals

who fall short to compete within the rigid occupational structure in Sri

Lanka. For examples, a school teacher who owes a tuition class at his

home may not pay any share of income from tuitions to the government

as the form of income tax. Similarly, a university student who doing a

part time job hiring an unlicensed three-wheeler at night and a drug

dealer negotiates a sale with his eventual customer on a street corner

are further examples for activities in SE. Therefore, giving definition

to and understanding the exact shape of SE is not an easy task to sort

out.

In general, there are two major approaches to define SE. They are the

‘Definitional Approach’ and the ‘Behavioural Approach’. According

to Definitional approach, Shadow Economic activities are simply in the

unrecorded economic activities while the behavioural approach is

focusing to its behavioural characteristics. Usually, SE is recognized

1 Well-functioning markets are embedded in a larger environment that creates the

necessary conditions for their effective operation. Markets do not function as well

without a certain degree of macroeconomic stability, trust and confidence.


116

as a part of economic activities involving goods and services which are

paid by cash, but not declared for tax. In addition to that, modern

thinking reveals that the formal and informal sectors in the economy

are depending on each other. Therefore, SE and formal economy could

be inter-dependent on each other.

A study conducted by Schneider and Enste (2000) in Sri Lanka, using

physical input-electricity consumption method 2 demonstrates that in

average 40% of Shadow economy was in 1989-90 periods. Another

study by Schneider (2004) has fund that an increasing trend of the size

of SE in between 1999-2003 periods and it has averaged at 45.9%.

However, recent study by Schneider et al. (2010) demonstrate a 43.9

% of average size of SE with decreasing trend for the periods from

1999 to 2007. In general, by 40% of economic activities are running

avoiding taxes and social security contributions; those will be greater

pessimistic barriers against the economic growth in a country.

The relationship between SE and the unemployment rate has long been

discussed by the researchers. According to Alanon and Antonio (2005),

higher unemployment rates encourage individuals to find a job in the

SE. In a study by Dell’Anno and Solomon (2008), have found a

positive relationship between unemployment rate and the SE.

Analysing the empirical literature, Davidescu and Dobre (2012) found

2 This method assumes that a part of the electricity supply is used for shadow

economy activities and that it is possible to calculate value added figures for the

shadow economy through concerning electricity consumption patterns. It compares

the dynamics of electricity consumption and the GDP. (See Nastav and Bojnec, 2005)


117

a strong evidence of uni-directional causality3. The findings of studies

which focus SE and level of unemployment may help the policy

makers to get rid of extensive barriers of growth and development

structures in Sri Lanka. Therefore, this study attempts to investigate

the appropriate models to estimate the SE as a percentage of GDP,

using unemployment rate and GDP index as proxies. Size of the SE in

Sri Lanka is not observed yet after 20074 and it is hard to find any study

which accomplishes to reveal the nature and relationship between

growth of SE and official estimates (GDP) in the economy.

LITERATURE REVIEW

Monitoring the characteristics and functions of SE is very important in

policy making point of view which directed towards improving social

and economic conditions of a country. Number of definitions to a

shadow economy can be available in literature at all the aspect of

political, sociological and economic. Considering the economic

definitions, Dell’Anno and Solomon (2008) classify shadow economy

into two categories as labour-oriented definitions and size-oriented

definitions.

3 Uni-directional causality is indicated if unemployment rate cause SE, then SE does

not cause unemployment rate. 4 The newest estimation for the Sri Lankan SE is observed through the study conduct

by Schneider et al. (2010). They found that Sri Lankan SE is 42.2 % share of GDP in

2007.


118

The Labour-oriented definitions focus on the impact of shadow

economy on the labour market while the Size-oriented definitions

consider the relationship between the state regulations and the

operation of business. This study uses the size-oriented classifications

of the SE in line with Dell’Anno and Solomon (2008). Shadow

Economic activities generate unofficial income5 in a country, and it

will create a problem for all the economic units in general and to the

government in particular in terms of policy planning and strategic

perspectives. Generally, the creation and improvements of a Shadow

Economy can possibility to be any forms of formal, informal and

illegal economic activities. The formal sector of the economy deals

5 Income generated through economic activities, which are not reported to the

governments and taxes are not paid.

Definitions of Shadow Economy

Labor oriented Size oriented

Figure 1: Classification of Shadow Economy

Economic Sociological Political

Source: Dell’Anno and Solomon, 2008


119

officially with the government paying taxes. The sector includes large

enterprises, the government, hospitals, universities, foreign investors

and etc. Increasing of Shadow Economic activities are badly affecting

in the base of formal economy. Rigidness of the rules and regulations

and the behavioural impacts of the government can induce the

widening of Shadow Economy.

The informal sector consists of small scale activities and they are too

small to be monitored. Workers and firms in this sector do not pay taxes

and hardly to be captured to the same laws as in the formal sector. The

informal sector in most of developing countries provides services,

manufactures, materials and etc. According to the Department of

Census and Statistics data base (2012), informal sector in Sri Lanka

accounts for about 57% to 62% of total employment. According to the

Consumer Finance and Socio-economic Survey (2012) the relative

share of informal economy was70%.

When we compare informal economy with SE, the shadow economic

activities do not pay the taxes to government. Yet, it includes illegal

economic activities other than the informal economy. However,

informal economy has a great link with shadow economy with slight

differences. According to the Organization of Economic Cooperation

and Development (2002), ‘Illegal Economic Activities’ are productive

activities that generate goods and services forbidden by law.

Furthermore, it can be unlawful services carried out by unauthorized

producers. When we compare particular illegal economy with shadow

economic activities, obviously illegal economic activities are one of


120

the important aspects in the Shadow Economy. It means that

improvements in illegal economic activities in a country will

essentially increase the size of the Shadow Economy.

Table 1: Types of Shadow Economy (Lippert and Walker, 1997

and Remarks by Schneider)

Type of

activities

Monetary transactions Non-monetary

transactions

Illegal

activities

Trade in stolen goods, drug

dealing and manufacturing,

prostitution, gambling,

smuggling, fraud, etc.

Barter of drugs, stolen

goods, smuggling, etc.:

(Production or growing of

drugs for own use, theft for

own use)

Tax evasion Tax

avoidance

Tax

evasion

Tax avoidance

Legal

activities

Unreported

income from

self-

employment,

wages,

salaries and

assets from

unreported

work related

to lawful

goods and

services

Employee

discounts,

fringe

benefits

Barter

of

lawful

goods

and

services

All do-it

yourself work

and

neighbourly

help

Source: Schneider, 2001

The current study does not include all the above categories in table 1

in modelling and analysing. It includes only the market based legal

productions of goods and services, similar to the definition given by

Schneider et al. (2010) that are purposely buried away from public

authorities due to any of the following reasons.


121

- To avoid paying taxes

- To avoid paying for social security contributions

- To avoid certain legal labour market standards

(Minimum wages, maximum working hours, etc.)

- To avoid fulfilling certain administrative procedures. (Questionnaires,

forms)

Alternative methods that can be applied to estimate the size of Shadow

Economy can available from literature. Schneider and Enste (2000)

classified those methods under three categories as direct, indirect and

modelling approaches. Majority of recent scholars use Multiple

Indicator Multiple Cause Model (MIMC) under Structural Equation

Modelling (SEM) approach to measure the size of the shadow

economy.

Schneider (2001) use currency demand approach and SEM to estimate

the size of the 18 Asian and 21 OECD countries. Then he investigate

that the shadow economy in terms of value added was 25.8% of official

GDP in the 18 Asian countries and 16.7% of official GDP in the 21

OECD countries.


122

Alanon and Antonio (2005) use SEM with latent variables to estimate

the size of the Shadow Economy in Spain for the 1976-2002 period.

They found in their study that a higher share of GDP (18%) in shadow

economy and it has been significantly influenced by three separate

factors: tax burden, the degree of regulation and unit labour costs. They

further found that a positive correlation between GDP, money demand

and the level of the Shadow Economy. In order to justify the results

from previous studies Dell’Anno et al. (2007) had measured the size of

Figure 2: Methods to Estimate Shadow Economy

Methods to estimate the Shadow Economy

Direct Indirect Model

The MIMIC Model

- Samples based on voluntary replies

- Well designed surveys

-Tax auditing

- The discrepancy between national expenditure and income

statistics

- The discrepancy between the official and actual labour force

- The transactions approach

- The currency demand approach

- The physical input / Electricity consumption method

- (The Kaufmann-Kaliberda method and The Lacko method)

Source: Schneider and Enste, 2000


123

the shadow economy in three Mediterranean countries: France, Spain

and Greece using the SEM approach. Unlike Alanon and Antonio

(2005), their results confirm that unemployment, the fiscal burden and

self-employment are the main causes of the shadow economy in these

countries. Moreover, they have identified that an inverse relationship

exists between the official GDP growth rate and the unofficial

economy.

In Asian experiences in measuring the size of the Shadow Economy

Chaudhri, Schneider and Chattopadhyay (2006) investigate the size of

the shadow economy in Indian states over the period 1974/75 to

1995/96 using the SEM approach. According to their analysis the

average growth rate of the SE has tended to decrease in the period after

the liberalization of the Indian economy in 1991/92. They show that

the growth in the size of the SE was approximately 4% in scheduled

election years relative to the normal years. Schneider (2004) estimates

the size of the SE for 145 countries all over the world using the SEM

approach over the period 1999 to 2003. In Sri Lanka, size of the SE

increased from 44.6% to 47.2% as a percentage of official GDP. In

context of Asia, it was 30.8% in 2002/03. In addition to that, another

empirical estimation of SE for 145 countries by Schneider (2004) from

1999 to 2003, verifies the increasing trend of SE in Sri Lanka.

However, Schneider et al. (2010) demonstrate a decreasing trend in SE

from 1999 to 2007 and it differs from the previous results regarding

the estimated sizes of SE in Sri Lanka.


124

In addition to the measuring the size of SE, scholars have attempted to

analyse the impact of SE on overall economy and the structural

relationship of factors affecting in growing SE sector in an economy

widening the dimensions of research area. In Dell’Anno and Solomon

(2008) estimate the size of US shadow economy using SEM to find

structural relationship exists between the SE and the unemployment

rate in the US. Furthermore, they extend the Okun’s law to estimate

the structural relationship between growth rate of official GDP, SE and

unemployment rate. They found a significant positive relationship

between SE and unemployment rate (UR).

Similarly, a strong positive relationship between SE and UR find by

Davidescu and Dobre (2012). They analysed the causal relationship

between UR and US Shadow Economy using SEM under Toda-

Yamamoto approach. On the base of empirical results they have

pointed out that a strong evidence of uni-directional causality running

from unemployment rate to shadow economy (at 1% level of

significance).

Schneider (2009) uses MIMIC approach to analyse the relationship

between corruption and the Shadow Economy and fund a positive

relationship between the SE and corruption. In analysing the

interdependent, The SE influences corruption more than corruptions to

SE. Biswas, Farzanegan and Thum (2012) observe how the SE affects

pollution and how that effect depends on corruption levels in public

administration. Their estimates verified that the size of SE and the level

of pollution depend on the level of corruptions. Dreher, Kotsogiannis


125

and McCorriston (2005) capture the relationship between institutional

quality, the SE and corruption. Their results show that an improvement

in institutional quality reduces the SE directly and corruption by both

directions: direct and indirect.

Illustrating the impact of unemployment rate to the dimensions of

Shadow Economy is a major interest field in recent studies in different

country context. According to Dell’Anno and Solomon (2008), the

empirical evidence suggests two important factors in discussing the

growth of SE. They are ‘reduction in official working hours’ and ‘the

influence of unemployment rate’. Therefore, most of researches to

estimate size of the SE use unemployment rate as a main causal

variable. Alanon and Antonio (2005), Dell’Anno et al. (2007),

Dell’Anno and Solomon (2008), Schneider (2009), Davidescu and

Dobre (2012) found unemployment as a highly significant causal

variable under the SEM approach. In more detail, the graphical

comparison made by Dell’Anno and Solomon (2008) can be used to

identify the way of significance of the unemployment rate.

Therefore, unemployment rate highly affects to the dimensions of SE

in USA. This result was pretty much same in the other European

countries too. In context of Sri Lanka, the estimations from Schneider

et al. (2010) and the behaviour of unemployment rates in certain

periods demonstrate a positive trend over the time period.


126

Figure 3: Unemployment vs. SE in United States

Source: Dell’Anno and Solomon, 2008

According to the literature, researchers has used different types of SEM

approaches to estimate the size of SE. Alanon and Antonio (2005) has

used Multiple Indicator Multiple Cause model (MIMIC) including the

variables such as tax burden, unit labour costs, waged employment

rate, unemployment rate, disposable income and public consumption

as their principle causes. They have used GDP, demand for money

(money in the hands of individuals) and energy consumption as

indicators to measure the SE. Their results are shown that disposable

income and public consumption are not statistically significant to use

as causal variables of the SE.

In Dell’Anno and Solomon (2008), have used nine causal variables to

estimate the size of the SE. They are aggregate index of tax burden,

personal current tax/GDP, taxes on production and import/GDP, taxes


127

on corporate income/GDP, contribution for government social

insurance/GDP, government unemployment insurance/GDP,

unemployment rate, self-employment/labour force and index of

bureaucracy. Indicators were the ratio of narrow money supply to

broad money monetary aggregate (M1/M2), Index of real GDP and

labour force participation rate. Also, Davidescu and Dobre (2012) use

the same causal and indicator variables in their research.

SEM approach with benchmark equation has been estimated by

Andreas and Schneider (2009) to examine the relationship between

corruption and the SE. He has used two unobserved indexes, SE and

corruption and causal and indicator variables. Business regulations,

unemployment rate, transfers and subsidies and government

consumption have been used as the causes of the SE and GDP growth,

Labour force participation and ratio of M0 to M1 for indicators.

Subsequently, he has used government effectiveness, fiscal freedom,

bureaucracy costs and rule of law as the causes of corruption and Real

GDP per capita, bribes and judicial independence as for indicators.

It is hardly to be found a specific study from Sri Lankan literature, with

the criteria of three model selections, calculation and growth

relationship analysis altogether. However, it is important to investigate

suitable models to estimate the Sri Lankan SE and to identify whether

which types of trend shown by the estimated values as a percentage of

official GDP. Three steps have to be followed. Initial portion of the

study is therefore intends to choose and develop suitable, fitted MIMIC

model or models, while using the structural relationship between the


128

unemployment rate and SE to estimate the size of the Shadow

Economy in Sri Lanka. Then the following portion is to estimate SE

for each fitted model through benchmark calculations. Finally, to

examine if estimated models and their calculations are accepted by re-

examination of Okun’s law and to contrast estimated results in growth

terms to the growth of official estimates (GDP) of the economy.

METHODOLOGY

The size of the Shadow Economy cannot be directly observed and

therefore, it demand a proper statistical approach to measure the size

of the economy. To fill the gap this study attempts to estimate the size

of the Shadow Economy by developing a Multiple Indicator Multiple

Cause (MIMIC) model under Structural Equation Modelling (SEM)

approach. Researchers use Structural Equation Models in order to

measure the relationships among unobserved and observed variables.

All variables have been tested for the unit root at levels of differences.

This method considers SE as an unobserved variable link with a set of

observed causal variables and also with a set of observed indicators to

reflect the changes in the size of the SE. These sorts of models include

two types of equations systems: the ‘structural equation’ and the

‘measurement equation’. An equation which explains the relationship

between unobserved variable (𝜂) and the causes (𝑋𝑛 ) is called the

‘Structural Equation’. If we assume the SE is linearly determines by a

set of observable exogenous causes ( 𝑋1,𝑋2, 𝑋3. . . 𝑋𝑛) and the error

term 𝜀𝑖 it can be expressed as;


129

𝜼 = 𝛽1(𝑋1)+ 𝛽2(𝑋2) + 𝛽3(𝑋3) + ……………… 𝛽𝑛(𝑋𝑛) + 𝜀𝑛 (1)

On the other hand, the equations that link indicators (𝑌𝑛 ) with the

unobserved variable (𝜂) is called the measurement model. Under thw

assumption of the Shadow Economy linearly determines subject to set

of observable exogenous indicators ( 𝑌1,𝑌2, 𝑌3….𝑌𝑛) and to error terms

( 𝜉1,𝜉2, 𝜉3….𝜉𝑛), it can be expressed as,

𝒀𝟏 = 𝛼1𝜂 + 𝜉𝑖, 𝒀𝟐 = 𝛼2𝜂 + 𝜉𝑖, 𝒀𝟑 = 𝛼3𝜂 + 𝜉𝑖……, 𝒀𝒊 = 𝛼𝑖𝜂 + 𝜉𝑖 (2)

Here the structural error term (𝜀𝑛 ) and measurement errors ( 𝜉1 ,𝜉2 ,

𝜉3 …. 𝜉𝑛 ) are normally distributed, mutually independent and all

variables are taken to have zero expectation in line with OLS properties.

This study depends only on secondary data and information sources. As

determinants of the existence of the shadow economy, the following

variables are included in this research. Six causal variables and three

indicators are going to be used to visualize the size of the Shadow

Economy in Sri Lanka using STATA-12 statistical software.


130

𝜶2

𝜷1

𝜷2

𝜷3

𝜷4

𝜷5

𝜶1

𝜶3

𝝃1

− 𝝃2

𝝃3

𝜼

𝜺n

𝜷6

𝝃2

Figure 4: The MIMIC Model Path Diagram with Variables and

Coefficients

Source: Authors Preparation

Each and every possibilities at least with three causal variables are run

in the software and identify the most fitted and suitable MIMIC models

according to following criterion.

1. The unemployment rate uses as a proxy to identify the fitted models.

Therefore, the causal variable unemployment rate ( 𝑋3 ) should be

significant with either positive or negative coefficient.

2. Model should be fitted under either 95% or 99% confidence level. (𝒳2

and RMSEA)

3. Lowest values of AIC and BIC are the final choice, if there are number

of fitted models in line with above three conditions.

Tax on Goods & Services / GDP × 100

(X1)

X1

Unemployment rate

(X3)

Index of Real GDP (Y1)

M1/M2 Ratio (Y2)

LFPR (Y3 )

Shadow Economy

Own Account Employment: / T: Emp: × 100 (X6)

Tax on Income & Profits / GDP × 100 (X2)

()X2

Pub: Emp: / T: Emp: × 100 (X4)

Pvt: Emp: / T: Emp: × 100 (X5)


131

After the structural analysis, we can identify the fitted model with

significant coefficients and their signs; which can be used to obtain the

size of SE. This study use a benchmark equation to estimate the size of

SE as a percentage of the GDP in Sri Lanka. It can be expressed as in

the form of equation 3:

[�̃�𝑡 /𝐺𝐷𝑃𝑏𝑎𝑠𝑒 ] × [𝜂∗𝑏𝑎𝑠𝑒

/𝐺𝐷𝑃𝑏𝑎𝑠𝑒] × [𝐺𝐷𝑃𝑏𝑎𝑠𝑒/�̃�𝑏𝑎𝑠𝑒 ] × [𝐺𝐷𝑃𝑏𝑎𝑠𝑒 /𝐺𝐷𝑃𝑡]

=[�̂�𝒕/𝑮𝑫𝑷𝒕] (3)

The equation 3 can be simplify as in equation 4,

[�̃�𝑡 × [𝜂∗𝑏𝑎𝑠𝑒

/�̃�𝑏𝑎𝑠𝑒 ] = �̂�𝒕 (4)

Where �̃�𝑡 for the value of structural calculation as a percentage of

GDP from the selected MIMIC model for year t, 𝜂∗𝑏𝑎𝑠𝑒

for the average

size of the previous estimations of Shadow Economy in the base year,

�̃�𝑏𝑎𝑠𝑒 for value of the structural calculation from the selected MIMIC

model for the base year and �̂�𝑡 for size of the SE as a percentage of

the Gross Domestic Production in Sri Lanka. Therefore, this study

requests a base year to estimate the size of SE in Sri Lanka. Then chose

year 2002 6 as the base year and use the average size of previous

estimates for Sri Lankan Shadow Economy.

In order to verify the estimated results for the SE in Sri Lanka and to

collaborate the relationship between the growth of official economy

and the growth of the SE, set of OLS (Ordinary Least Squares)

6 This year reports two previous estimations for the average size of Sri Lankan SE

and also a popular for number of national estimates as the base year.


132

regressions are used to re-examine the well-known Okun’s law for

each estimated MIMIC models.

Table 2: Estimates of the size of Sri Lankan Shadow Economy in

2002

Author/Authors Source/method Size of Shadow

Economy

Schneider (2004) MIMIC Model 47.2%*

Schneider et al.

(2010) MIMIC Model 44.1%

Average size 45.65%

Note: *(Mean of 2002/3)

Then the benchmark equation can be written as below,

[�̃�𝑡 × [𝜂∗2002

/�̃�2002 ] = �̂�𝒕 (5)

According to the Okun’s law, there is an inverse relationship exists

between unemployment rate and output growth in an economy.

Therefore the current study will use equation 6 as an initial equation

for Okun’s law to investigate the structural relationship.

𝒈𝒕𝒚 = ∝1 Δ𝑢𝑡+ 𝑔𝑡

𝑛𝑖+ 𝜀𝑡 (6)

Where Δ𝑢𝑡 for the change in unemployment rate, 𝑔𝑡𝑛𝑖 for the annual

growth rate of the estimated SE of ith model and 𝑔𝑡𝑦

for the annual

growth rate of the official economy.

The study collected the Sri Lankan annual time series data for the

period from 1990 – 2012 to analyse and investigate the nature of

relationships exists between causes or indicators and the shadow


133

economic activities. However this is not a typical time series analysis,

and uses special methodology named Structural Equation Modelling to

measure the impact of ‘latent’ or unobserved variables.


Test for the Unit-root

The variables used in the estimation are defined in table 2 and all of

them have been tested for unit root in levels or differences. Augmented

Dickey-Fuller (ADF) test have been used for making data stationary at

I (0) or I (1) levels. The summary of the unit root test is demonstrated

in the table 3.

Table 3: Summary for ADF test results

Variable Level 1st Difference

Tax on Goods and Services -0.457 -3.852**

Tax on net income and profits -2.592* -5.352***

Unemployment rate -1.546 -4.429***

Public employment -2.671* -3.753**

Private employment -4.413*** -

Own account workers -1.579 -9.719***

Real GDP Index -1.054 -5.344***

M1/M2 Ratio -1.381 -3.333**

LFPR -3,255** -

Note: t statistics are given in table and *10%, **5% and ***1%

significance levels.



134

As it is clear from the Table 3, the level of private employment and

Labour Force Participation Ratio (LFPR) are the significant variables

(at I(0)). Other variables have only the significant at I(1) level. Then

the set of estimated MIMIC models are shown in table 4.


135

Table 4: Estimated Coefficients of the MIMIC models and Descriptive Statistics

Mo

del

s

Tax

es o

n

Go

od

an

d

Ser

vic

es

Tax

es o

n

inco

me

and

pro

fits

Un

emp

.

Rat

e

Pu

bli

c

emp

.

Pri

vat

e

emp

.

Ow

n

acco

un

t

wo

rker

s

GD

PI

M1

/M2

LF

PR

𝒳2

(p-v

alu

e)

RM

SE

A (

p-

val

ue)

AIC

BIC

Df

X1 X2 X3 X4 X5 X6 Y1 Y2 Y3

MIMIC

3-1-3a

- - 0.006

(- .01)

- 0.718**

(- 2.76)

- 0.48***

(23.76)

- - 0.41

- 0.12

(- 0.24)

- 0.22***

(- 6.14)

22.91++

0.285++

193.38

214.95

08

MIMIC

3-1-3b

- - 0.24*

(1.74)

- 0.9***

(3.95)

- 0.41***

(- 4.67)

- 0.447

- 0.036

(- 0.56)

- 0.23

1513.6++

2.437++

1754.3

1772.5

11

MIMIC

4-1-3a

- - 0.42*

(- .79)

0.75***

(5.22)

0.367

(1.33)

- 0.23***

(- 4.36)

- - 0.187

(- 0.69)

0.265

(1.2)

0.63***

(8.8)

30.07++

0.275++

272.3

299.6

11

MIMIC

4-1-3b

0.99***

(5.03)

- 0.29**

(2.69)

0.31

(0.84)

- 0.34

(0.99)

- 0.34*

(- 1.65)

- 0.05

(- 0.22)

0.526***

(4.16)

14.09

0.133

254.5

282.8

10

MIMIC

4-1-3c

- - 0.47*

(- .74)

0.75***

(3.82)

- - 0.24

(- 0.68)

0.06

(0.22)

- 0.21

(- 0.68)

0.22

(0.91)

0.62

(6.61)

26.44++

0.267++

279.6

307.9

10

MIMIC

4-1-3d

0.71**

(2.45)

- 0.61*

(1.68)

- - 0.12***

(- 3.6)

0.1

(0.42)

- 0.24

(- 0.66)

.003

(0.01)

0.67***

(4.32)

18.58+

0.193

304.7

333.1

10


136

Mo

del

s

Tax

es o

n

Go

od

an

d

Ser

vic

es

Tax

es o

n

inco

me

and

pro

fits

Un

emp

.

Rat

e

Pu

bli

c

emp

.

Pri

vat

e

emp

.

Ow

n

acco

un

t

wo

rker

s

GD

PI

M1

/M2

LF

PR

𝒳2

(p-v

alu

e)

RM

SE

A (

p-

val

ue)

AIC

BIC

Df

MIMIC

5-1-3a

0.72**

(2.19)

- 0.39*

(- .79)

0.62***

(3.83)

0.27

(1.16)

- 0.12

(- 0.3)

- - 0.308

(- 1.01)

0.13

(0.44)

0.71***

(9.41)

31.16++

0.263++

309.7

346.1

12

MIMIC

5-1-3b

0.82***

(4.11)

- 0.56***

(3.6)

0.29

(1.1)

- 0.11***

(- 3.87)

0.19

(0.77)

- 0.235

(- 0.91)

- .009

(- 0.04)

0.67***

(7.73)

18.64

0.137

371.78

406.98

13

MIMIC

6-1-2a

0.92 ***

(4.44)

- 0.3

(- .16)

0.29**

(2.09)

0.38

(1.33)

- 0.1***

(- 3.14)

0.133

(0.44)

- 0.123

(- 0.23)

- 0.65***

(5.62)

14.25

0.184

513.3

554.2

08

MIMIC

5-1-2a

- 0.42***

(- 4.34)

- - 0.36***

(- 4.52)

- 0.37**

(- .09)

0.11**

(2.42)

0.18

(1.00)

0.047***

(5.76)

1

- 56.85++

0.515++

341.26

371.92

08

X1 X2 X3 X4 X5 X6 Y1 Y2 Y3

MIMIC

4-1-2a

1.04***

(6.13)

- 0.3

(-0.96)

0.258***

(2.63)

0.295

(0.85)

- - - 0.359*

(- 1.75)

- 0.51***

(3.93)

9.68

0.2

308.96

333.94

05

MIMIC

4-1-2b

- - 0.691**

(2.61)

0.299

(0.87)

- 0.467**

(- 2.41)

0.11

(0.4)

0.056**

(2.47)

0.64***

(4.1)

- 21.7++

0.381++

268.68

293.66

05


137

Mo

del

s

Tax

es o

n

Go

od

an

d

Ser

vic

es

Tax

es o

n

inco

me

and

pro

fits

Un

emp

.

Rat

e

Pu

bli

c em

p.

Pri

vat

e

emp

.

Ow

n

acco

un

t

wo

rker

s

GD

PI

M1

/M2

LF

PR

𝒳2

(p-v

alu

e)

RM

SE

A (

p-

val

ue)

AIC

BIC

Df

MIMIC

3-1-2a

- - - 0.71**

(- 2.14)

- 0.24

(- 0.55)

0.45***

(3.43)

- - 0.4***

(- 3.92)

- - 0.115**

(- 2.32)

7.12

0.184

373.2

391.3

04

MIMIC

3-1-2b

1.07***

(10.41)

- 0.32***

(3.04)

0.197

(0.55)

- - - 0.335**

(- 2.27)

- 0.49***

(4.94)

8.34+

0.278+

295.67

314.97

03

Notes: z – statistics are given in parentheses for each coefficient. Coefficients are significant if | z - statistic | > 1.96 for 95% confidence.

*** Means significance of coefficients under 99% of confidence level. ** For 95% and * for 90 % respectively. ++ Means good fitting (p-value > 0.01) where 99% confidence and + Means good fitting (p-value > 0.05) where 95% confidence.

RMSEA – Root mean squared error of approximation. P-value for test of close fit (RMSEA > 0.05)

AIC – Akaike’s information criterion, BIC- Bayesian information criterion

Df- Degrees of freedom. (Values obtain from the each estimated models)

Model selection Criteria:

Unemployment rate (X3) should be significant – Coefficient can be either positive or negative. (Assumption -Theoretical)

Model should be fitted under either 95% or 99% confidence level. (𝒳2 and RMSEA)

Lowest values of AIC and BIC are the final choice if there are number of fitted models in line with above three conditions.



138

Estimated MIMIC Models

MIMIC 5-1-2a, MIMIC 4-1-2b and MIMIC 3-1-3a are three

estimated models which have been selected according to the model

selection criterion. Then these three models can be used to extract the

structural equations using estimated coefficients. The Structural

Equation (7) is extracted by the coefficients from MIMIC 5-1-2a.

�̃�𝑡 /𝐺𝐷𝑃2002 = - 0.42 X1t - 0.36 X3t - 0.37 X4t + 0.11 X5t (7)

(- 4.34) (- 4.52) (- 2.09) (2.42)

According to MIMIC 5-1-2a, the shadow economy of Sri Lanka as a

percentage of GDP will depend on tax on domestic goods and service,

unemployment rate, public employment and private employment.

Here, three coefficients except public employment negatively affected

to the size of the Shadow Economy in Sri Lanka. The Structural

Equation (8) is extracted by the coefficients from MIMIC 4-1-2b.

�̃�𝑡 /𝐺𝐷𝑃2002 = 0.69 X3t - 0.467 X5t (8)

(2.61) (- 2.41)

According to MIMIC 4-1-2b, Shadow Economy, as a percentage of

GDP, will depend only on the unemployment rate and private

employment. Here, unemployment rate positively affect to the size of

the Shadow economy in Sri Lanka while the private employment

demonstrate an inverse relationship. The Equation (9) is extracted by

the coefficients from MIMIC 3-1-3a.


139

�̃�𝑡 /𝐺𝐷𝑃2002 = - 0.718 X3t + 0.48 X5t (9)

(- 2.76) (23.76)

According to MIMIC 3-1-3a, shadow economy, as a percentage of

GDP will depend on only unemployment rate and privet employment.

Here, the unemployment rate negatively affects to the size of the SE

while the private employment demonstrate a positive relationship.

Then these three structural equations are used to perform the

benchmark calculations.

Benchmark Calculations

Estimate coefficients from each estimated model and average base year

value for the Shadow economy will be used to measure the average

size of shadow economy for Sri Lanka. Benchmark calculations are

performed separately for each model7.

Table 5: New Estimate for Sri Lankan Shadow Economy (1990-

1997)

Estimates of

SE (% 0f

GDP)

Year

1990 1991 1992 1993 1994 1995 1996 1997

MIMIC

5-1-2a

91 86.3 78.8 70 64.86 63.5 57.29 55.7

MIMIC

4-1-2b

14.79 25.77 26.7 32.47 36.15 37.86 42.18 41.72

MIMIC

3-1-3a

14.45 25.51 26.45 32.27 35.98 37.73 42.1 41.65

7 All benchmark calculations for relative sizes of SE are shown in appendix


140

Table 6: New Estimate for Sri Lankan Shadow Economy

(1998-2004)

Table 7: New Estimate for Sri Lankan Shadow Economy

(2005-2012)

Source: Authors Calculations

According to MIMIC 5-1-2a, Sri Lankan SE varies between 91% and

32% of GDP and demonstrates a decreasing trend over the period under

study. On the other hand, according to MIMIC 4-1-2b and MIMIC 3-

1-3a, the size of the SE varies between 14% and 52% of GDP and

demonstrates an increasing trend. Moreover, the estimated values from

MIMIC 4-1-2b and MIMIC 3-1-3a are showing a similar observation

for all respective years. Therefore, it is worthy to discuss the results

from each model separately with concerning the effects from causal

Estimates of

SE (% 0f

GDP)

Year

1998 1999 2000 2001 2002 2003 2004

MIMIC

5-1-2a

51.1 49.53 45.64 46.1 45.65 42.88 42.18

MIMIC

4-1-2b

40 43.4 45.9 47.86 45.65 46.36 49.47

MIMIC

3-1-3a

39.97 43.4 45.93 47.9 45.65 46.37 49.5

Estimates

of SE (%

0f GDP)

Year

2005 2006 2007 2008 2009 2010 2011 2012

MIMIC

5-1-2a

42.08 41.09 38.23 38.39 36.07 33.45 32.26 33.62

MIMIC

4-1-2b

50.32 47.1 49 48.15 48.38 49.22 49.7 51.3

MIMIC

3-1-3a

50.37 47.16 49.13 48.25 48.47 49.34 49.84 51.45


141

variables including in each model. The OLS regression analysis for

Okun’s law and augmented equations will help to recognize the

relevance of three MIMIC models and the fundamental theoretical

basis of Okun’s law.

Okun’s Law and Augmented Equations

To examine the negative relationship between the change in

unemployment and the growth rate of official output (GDP) in Sri

Lanka within the respective study period (1990-2012), this study

persuades a simple OLS regression analysis between the annual growth

of GDP and the change in unemployment rate.

𝑔𝑡𝑦

= ∝1 ∆ 𝑢𝑡 + 𝜀𝑡 (10)

𝑔𝑡𝑦

= Annual growth of official GDP

∆ 𝑢𝑡 = Change in unemployment rate

Results satisfy the expected negative relationship and help for the re-

examination of Okun’s law with augmented versions of equations for

each MIMIC model.


142

Figure 5: Relationship between GDP Growth Rate and Change of

Unemployment (1990-2012)

Source: Authors preparation using MINITAB 13 statistical software

Then three separate OLS regressions were run on each MIMIC model

to recognize the nature and relationship between growth in official

GDP and change in the unemployment rate.

𝑔𝑡𝑦

= ∝1 ∆ 𝑢𝑡 + 𝑔𝑡𝜂1

+ 𝜀𝑡 (11)

𝑔𝑡𝑦

= ∝1 ∆ 𝑢𝑡 + 𝑔𝑡𝜂2

+ 𝜀𝑡 (12)

𝑔𝑡𝑦

= ∝1 ∆ 𝑢𝑡 + 𝑔𝑡𝜂3

+ 𝜀𝑡 (13)

Where 𝑔𝑡𝑦

for annual growth of official GDP, ∆ 𝑢𝑡 for change in

unemployment rate and 𝑔𝑡𝜂1

, 𝑔𝑡𝜂2

, 𝑔𝑡𝜂3

for annual growth rate of SE

for MIMIC 5-1-2a, MIMIC 4-1-2b and MIMIC 3-1-3a. Summary

of estimated results were shown in table 8.

Annual growth rate of GDP

Ann

ual c

hang

e of

Une

mpl

oym

ent ra

te

7.55.02.50.0

1.0

0.5

0.0

-0.5

-1.0

-1.5


143

Table 8: Summery Statistics for the OLS Regressions

The results in table 8 very clearly shows that initial equation for Okun’s

law and all three models (Augmented equations) are statistically

significant at 95% confidence level. Though, all equations show lower

R2 values. This may because of purposive chosen of some variables

and fitted into the regression models. Re-examination of Okun’s law

for MIMIC 5-1-2a, MIMIC 4-1-2b and MIMIC 3-1-3a demonstrate

the expected negative relationship between GDP growth rate and

change of unemployment rate. However, most importantly any of three

models do not illustrate a significant relationship between growth of

estimated SE and growth of official GDP.

Test Model GDP

growth and

∆ Unem:

Rate

GDP

growth

and

growth of

S:E:

F-

prob:

R2

Okun’s

law - -1.77**

(-2.63) - 0.016** 0.2194

Augment

ed Eq: 1

MIMIC

5-1-2a

-1 94**

(-2.65)

0.104

(0.65)

0.048** 0.2726

Augment

ed Eq: 2

MIMIC

4-1-2b

-1.996***

(-2.95)

- 0.187

(- 1.42)

0.023** 0.3280

Augment

ed Eq: 3

MIMIC

3-1-3a

-2***

(-2.95)

-0.186

(- 1.42)

0.023** 0.3281

Note: t statistics are given in the parenthesis.

Coefficients are significant if | t-statistic | > 1.96 for 95% confidence.


144

OVERALL DISCUSSION OF RESULTS

This section will broadly discuss the whole set of results coming under

each chosen MIMIC model and their empirical validity; theoretical

support by previous studies and the ground realities in the Sri Lankan

economy.

MIMIC 5-1-2a

Benchmark calculations for MIMIC 5-1-2a demonstrate a decreasing

trend of estimated SE as a percentage of Sri Lankan GDP over 1990-

2012.

Figure 6: Estimated SE from 1990-2012 According to MIMIC

5-1-2a


Four causal variables have been used to estimate the percentage

demonstrated on the vertical axis in figure 6. Tax on domestic goods

and services demonstrates an inverse relationship to the size of the SE.

Calculated coefficient between two variables was at 0.42. It is evident

to suggest that higher government tax revenue on domestic goods and

Year

Sha

dow

Eco

nom

y as

a %

of GDP

201220102008200620042002200019981996199419921990

90

80

70

60

50

40

30


145

services will contract the size of the SE. Therefore, this is not only

about the increase of tax per unit of commodity or service, but also

about the number of economic activities which counts under the

government tax scheme. According to the literature, Schneider et al.

(2010) also estimated Sri Lankan SE with a decreasing trend from

1999-2007. Share of direct taxation and total tax burden are two

variables they use in their MIMIC model specifications. Therefore, the

negative effect from government tax revenue on domestic goods and

services will provide a hint, where Sri Lankan tax collection

procedures getting smoother and it covers lots of goods and services

over the time. On the other hand, tax on net income and profits is

insignificant and not much important as a causal variable to explain the

SE within the results; because more than 70% of government tax

revenue comes from indirect taxes (through goods and services).

Unemployment rate, public employment and private employment are

the other significant causal variables in this model. The unemployment

rate also shows an inverse relationship with 0.36 coefficient. It means

that higher the unemployment rate will lower the size of SE in Sri

Lanka. This is confirming the empirical evidence (e g. Sri Lankan

unemployment rate shows a decreasing trend over time).


146

Figure 7: Estimated SE for MIMIC 5-1-2a and Unemployment

Rate in Sri Lanka


Figure 7 illustrates a positive behavioural pattern between average size

of SE and the unemployment rate of the economy. The under-

employment issue will be a major reason for this kind of dilemma

between estimated results and the ground reality. The mismatch

between wages and qualifications will generate under-employment

issue and the results can stimulate public sector employers to spend

less time for their job and spend more with Shadow Economic

activities.

Public employment demonstrates similar inverse relationship to the

size of the SE with 0.37 coefficient. Accordingly it can be conclude

that the higher the public employment in Sri Lankan economy will lead

to reduce the size of SE. This can persuade due to the higher degree of

job security and sticky wage rates within the public sector occupations.

On the other hand, private employment demonstrates positive

Year

201020062002199819941990

90

80

70

60

50

40

30

201020062002199819941990

17.5

15.0

12.5

10.0

7.5

5.0

MIMIC 5-1-2a UR


147

relationship to the size of the SE with 0.11 of coefficient, means that

when higher the private employment in Sri Lankan economy will tend

to higher the size of the SE. Uncertainty and dynamic wage patterns

with easy access for employer to make their own decisions can be some

reasons for this kind of result.

Figure 8: Public Employment (PE) and Privet Employment

(PVTE)


The augmented Okun’s law analysis for this model demonstrates a

significant inverse relationship between growth of GDP and change in

unemployment rate with 1.94 of coefficient. Moreover, there is no

significant relationship between growth of GDP and SE. This

insignificance might be due to Sri Lankan SE for not being a separate

part of official economic activities and it over-valuate and de-valuate

the official estimates.

Year

201020062002199819941990

24

22

20

18

16

14

12

201020062002199819941990

47.5

45.0

42.5

40.0

37.5

35.0

Public Employment Private Employment


148

MIMIC 4-1-2b

Benchmark calculations for MIMIC 4-1-2b demonstrate an increasing


2012.

Figure 9: Estimated SE from 1990-2012 According to MIMIC 4-

1-2b


There are two causal variables which were used to estimate above

percentages. Unemployment rate demonstrates a positive relationship

with the size of SE and persuade 0.69 of coefficient. Therefore, higher

the unemployment rate will leads to higher the size of SE in Sri Lanka.

This is in line with the empirical relationship demonstrated in figure

4.3 and accepts the mathematical proven by Dell’ Anno and Solomon

(2008). On the other hand, private employment demonstrates an

inverse relationship to the size of the SE with 0.467 of coefficient.

Therefore, it can be concluded that higher the private employment in

Year

Sha

dow

Eco

nom

y as

a %

of GDP

201220102008200620042002200019981996199419921990

50

40

30

20

10


149

Sri Lankan economy will tend to lower the size of the SE. This shows

different result to the argument develops through MIMIC 5-1-2a.

MIMIC 3-1-3a

Benchmark calculations for MIMIC 3-1-3a demonstrate an increasing


2012.

Figure 10: Estimated SE from 1990-2012 According to MIMIC

3-1-3a


There are two causal variables such as MIMIC 4-1-2b to estimate

above percentages. They are the unemployment rate and the private

employment. The unemployment rate demonstrates an inverse

relationship with the size of SE and persuade 0.72 of coefficient. It

means that higher unemployment rate will lower the size of SE in Sri

Year

Sha

dow

Eco

nom

y as

a %

of GDP

201220102008200620042002200019981996199419921990

50

40

30

20

10


150

Lanka. This is in line with the argument build through the results from

MIMIC 5-1-2a. On the other hand, the private employment

demonstrates a positive relationship to the size of the SE with of 0.48

coefficient. Therefore, higher the private employment in Sri Lankan

economy will tend to higher the size of the SE. This also supports the

similar argument that built through the results reached to MIMIC 5-1-

2a.

CONCLUSION

The purpose of the current study was to estimate the size of Sri Lankan

SE as a percentage of GDP and to estimate the five causal variables

and two indicators visualized in the existence SE of the Sri Lanka. With

this objective, initial part of the analysis has been attempted to choose

appropriate and fitted models through Structural Equation Modelling.

To obtain virtuous results for the structural analysis, data series have

been tested for the unit root using ADF test and three empirical models

were estimated. Most obvious finding found in this study is to an

inverse effect from tax on domestic goods and services, unemployment

rate and public employment to the SE. Secondly the study found that

the a positive effect of private employment to the SE sector.

Furthermore, tax on net income and profits is not strong enough to

explain the behaviour of SE. On the other hand, SE indicates a positive

relationship to the indicators Real GDP Index (GDPI) and M1/M2 ratio.

After introducing the four causal variables and two indicators to the

models, the size of SE in the country positively correlates with the


151

unemployment rate and negatively to private employment. Results

illustrate positive effect from and to. Moreover, public employment

and own account workers are not strong enough to explain the

behaviour of Sri Lankan SE. Overall results of the study reveals that a

less relationship between growth of estimated SE and growth of

official GDP. Consequently, it indicates that the growth of Shadow

Economy and official economy are not interdependent. In this context

policy makers should have paid their attention to adopt a more pro-

active policy package to increase the employment rate in both the

sectors of public sector and the private sector.

REFERENCEs

Alanon, A. and G.M. Antonio, 2005. Estimating the size of the shadow

economy in Spain: a structural model with latent variables.

Applied Economics, 37: 1011-1025.

Andreas, B. and F. Schneider, 2009. Corruption and the shadow

economy: A Structural Equation Model approach. IZA

Discussion paper, 4182: 1-39.

Biswas, A.K., M.R. Farzanegan and M. Thum, 2012. Pollution,

Shadow economy and corruption: Theory and evidence.

Ecological Economics, 75: 114-125.

Bollen, K.A., 1989. Structural Equations with latent variables. Wiley-

Interscience Publication, New York.

Central Bank, 2012. Sri Lanka Socio Economic Data. Central Bank of

Sri Lanka, Colombo. XXXV.

Central Bank. Various years, Annual Reports. Central Bank of Sri

Lanka, Colombo.


152

Chandrasiri, S., 2008. The labour market in Sri Lanka with special

reference to the informal economy. Sri Lankan economic

journal, 9(1): 113-141.

Chaudhuri, K., F. Schneider and S. Chattopadhyay, 2006. The size and

development of the shadow economy: An Empirical

investigation from States of India. Journal of Development

Economics, 80(2): 1-23.

Davidescu, A.A. and I. Dobre, 2012. The Causal relationship between

the unemployment rate and U.S. shadow economy. A Toda-

Yamamoto approach. Journal of Social and Economic

Statistics, 1(1): 21-34.

Dell’Anno, R., M.G. Antonio and A.A. Pardo, 2007. The shadow

economy in three Mediterranean countries: France, Spain and

Greece. MIMIC approach. Empirical Economics, 1-30.

Dell’Anno, R., and O.H. Solomon, 2008. Shadow economy and

unemployment rate in USA: is there a structural relationship?

An empirical analysis. Applied Economics, 40: 2537-2555.

Fleming, M.H., J. Roman and G. Farrell, Spring 2000. The Shadow

Economy. Journal of International Affairs, 53(2): 387-409.

Giles, E.A. and L.M. Tedds, 2002. Comments on taxes and the

Canadian underground economy. Canadian tax journal, 50(5):

1639-40.

International Labour Organization, 2013. Global Employment Trends.

International Labour Office, Geneva. Available at

http://www.ilo.org/wcmsp5/groups/public/---dgreports/---

dcomm/---publ/documents/publication/wcms_202326.pdf.

Labour Force Survey, 2009 and 2012. Statistical Appendix. Annual

reports, Department of Census and Statistics, Colombo.

Muller, P., G. Conlon, M. Lewis and I. Mantovani, 2013. From

Shadow Formal Economy: Levelling the Playing Field in the

Single Market. A study sponsored by the policy department A:

economic and scientific policy, European parliament, European

Union. Available at http://www.europarl.europa.eu/studies.

http://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/---

http://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/---


153

Nastav, B. and S. Bojnec, 2005. Shadow economy in Slovenia. Paper

presented at the 6th international conference of the faculty of

management Koper Congress Centre Bernardin, Slovenia, and

November: 24-26.

Ruge, M., 2010. Determinants and Size of the Shadow Economy-A

structural Equation Model. International Economic Journal,

24(4): 511-523.

Schneider, F., A. Buehn and C.E. Montenegro, 2010. New estimates

for the shadow economies all over the world, 24(4): 443-461.

Schneider, F. and D. Enste, 2000. Shadow economies around the

world: Size, Causes and consequences. IMF Working paper

series, WP/00/26. Available at

http://www.imf.org/external/pubs/ft/wp/2000/wp0026.pdf.

Schneider, F., 2001. The size and development of the shadow

economies and Shadow economy labour force of 18 Asian and

21 OECD countries. Invited paper presented for the Asia

Pacific Finance Association (APFA) Conference held in

Bangkok, Thailand, July, 22-25.

Schneider, F., 2004. The size of the shadow economies of 145

Countries all over the world: First result over the period 1999

to 2003. IZA Discussion paper, 1431. Available at

http://ftp.iza.org/dp1431.pdf.

http://ftp.iza.org/dp1431.pdf


154

155

PERSPECTIVES


156

IS ECONOMICS INDEPENDENT PERSPECTIVES

OF MORALITY? Sri Lanka Journal of

Economic Research

Volume 3 (1)

June 2015: 157-166

Sri Lanka Forum of


Danny Atapattu SLJER

Emeritus Professor, Department of Economics, University of Ruhuna, Sri Lanka

___________________________________________________________

157

On questions of morality, ‘Contemporary Economics’ stands mute.

Economists prefer sidestep moral issues. They like to say that they

study trade-offs, and incentives and interactions, leaving value

judgments to the political process and society. Today, Economics is

commonly regarded as separate from and independent of ‘Ethics.’ It is

not that the two are different subjects, like Mathematics and History,

but that they are disciplines unrelated in any fundamental way; because

they are seen to be so different from one another. Keeping economics

and ethics separate, is viewed as a highly desirable state of affairs. On

the one hand, their argument is that Economics is a science, and as such

is concerned with facts. Ethics, on the other hand, is concerned with

values, which are not amenable to the same rigorous treatment as facts.

With this view of facts and values, economists quite naturally want to

put as much distance as possible between ‘Ethics’ and Economics.

Economics did not begin as a science independent of Ethics. Adam

Smith, who is generally regarded as to be the founder of “Free Market

Economics,” was both a moral philosopher and an economist. Smith

came to his philosophy of economic behaviour, described in The


158

Wealth of Nations through his view of moral behaviour espoused in his

first book, The Theory of Moral Sentiments. He made no attempt to

draw a sharp line between ‘Ethics’ and Economics, or to keep the two

separate, Indeed, The Wealth of Nations is Smith’s attempt to explain

and defend a system of natural liberty (Smith, 1976a and 1976b).

Smith suggested that rational self-interest informed by moral

judgments based on fairness and justice, would lead to promote the best

interests of society, guided by the invisible hand of the market place.

He approached Economics with a deep sense of right and wrong and

talked about the need to maintain the ‘Laws of Justice’ in all things.

Efficiency was important so was sustainability. With freedom came

responsibility. Freedom did not mean that we could ravage the planet

for short term gain. It did not mean that mortgage salesmen could sell

loans which would force borrower into destitution. ‘Freedom’ in the

enlightenment sense, always carried a responsibility to others and to

society. Smith also believed that wealth should not just benefit the

individuals who created it, but wider society too. So he thought that the

rich should be taxed more than the poor. He also said that profit should

not be too high. If it were, he said, ruin would soon follow.

In Smith’s Sort of Economics, profit was defined as “returns sufficient

to maintain a business long term.” If they got too high, he believed that

the invisible hand would ensure a stiff dose of competition brought

them back into line. This was everyone’s interest because excessive

profits, act against social harmony. They increase income inequality.


159

Today, we have abandoned Smith’s Sort of Economics. In the decision

making process, there is no place for ‘ethical decision making’ in the

current context of Economics. Milton Friedman postulated that “the

social responsibility of business is to increase its profits.”

“Maximizing shareholder value” is how it would be described today.

Milton Friedman argued that competition between big businesses

suffices to safeguard the public interest, but in practice, it is almost

always insufficient, especially where there is collusion among the

players to safeguard their market dominance, and their political

influence (Friedman, 1962 and 1980).

We can observe that in the technology and many other sectors,

companies generate huge profits. Moreover, many of the financial

institutions that have made the modern market economies so unstable

in the recent years, make their profits from speculating. They just do

not make excessive profits, they hardly add any economic value either.

The same is true of the trading houses, commodities brokers, property

developers. Some of the big investment banks regularly make more

than $ 100 million a day, simply betting on the market. They gamble

but fiddle the outcome in their favour. Is it acceptable that many of the

largest companies in the world make their profits from gambling, rather

than supporting economic growth or financing business development?

It should be evident however, that this is not a reliable route to a good

society. Five major problems emerge around this strategy of profit

maximizing behaviour: instability, insecurity, inequality, monopoly


160

power and unsustainability. The problem of economic instability was

all too obvious in the wake of the global financial crisis. Economic

institutions pursuing profit without social responsibility resulted the

most sharply pronounced financial crash, since the Great Depression

in 1929.

Economics has made its own attempt to solve some of the problems

involved in the moral judgment, in what we know as Welfare

Economics. I believe this attempt has been a failure. ‘Welfare

Economics’ attempts to ask the question “What do we mean when we

say that one state of a social system is better than another in strictly

economic terms?” The Pareto-criterion compares two social states, A

and B, and makes a claim about whether the policy change that brings

us from A to B is desirable or not. The most celebrated answer given

is the Paretian optimum, which states that if in B all individuals have

the same welfare as in A, and at least one of them has a higher level,

then moving from A to B is a Pareto –improvement, and the Pareto

criterion recommends the move from A to B on grounds of efficiency.

Many, if not most, economists accept the Paretian optimum as almost

self-evident. Nevertheless, it rests on an extremely shaky foundation of

ethical propositions. As many have argued, the Pareto-criterion

remains undecided about the fairness or legitimacy of A as the starting

point. The Pareto-criterion also doesn’t attach any importance to

distributive issues in either A or B. If those who are living in misery

stay equally miserable, but due to some policy or social change the

ultra-rich become even richer, then can we all agree that this social


161

change is a social improvement? I am using this example of the Pareto

criterion merely to support the claim that economists need to think

more about the values embedded in their theories.

There is good reason to think that economic theory is itself in crisis and

that arriving at a better understanding of Economics, is a necessary

goal of long-term prosperity. Human societies neither move like

planets, nor like molecules or DNA, of which the movements can be

mapped with mathematical precision. The people in societies move

under the laws of motion determined by themselves. The ambition of

modern economic theory to resemble the Natural Sciences, has

excluded the aspects of economic life that have no reliable parallel in

nature, and therefore, cannot be understood with the aid of

methodologies designed for the Natural Sciences. The exclusion of the

moral quality of economic phenomena in the name of scientific

objectivity has produced a ‘reductionist’ and ‘determinist’ discipline.

A further cause of disquiet is the Mathematization of human behaviour

and desire for predictive models. Heavy emphasis on Mathematics has

caused Neo-classical Economics to rely upon increasingly

questionable presuppositions about individual autonomy and full

information in decision making. The over-reliance upon mathematical

formulae is at the root of why economists have become detached from

ethical aspects in their professional studies. The use of Mathematics in

Economics is undoubtedly very impressive, but there would appear to

be a paradox here. Mathematics being synonymous with rigour and


162

precision, how is it that, it plays such a role in a discipline where

vagueness reigns. As the economic and social world is so difficult to

grasp schematically, to reduce to simple laws, the temptation is to take

refuge in fictitious worlds, in models which have little to do with what

can actually be observed but which lend themselves to endless

mathematical refinement. The most important of life’s criteria and its

purpose are ignored. Behind the shield of Mathematical Formalism,

Economics has given remarkably little attention to its hidden moral

assumptions.

Further, we have created a society in which Materialism overwhelms

moral commitment, in which the rapid growth that we have achieved

is not sustainable environmentally or socially, if we do not act together

to address our common needs. Market Fundamentalism has eroded any

sense of community and has led to rampant exploitation of innocent

and unprotected individuals.

I believe that much of Economics, as it is now taught, is morally

bankrupt. The renowned economist John Maynard Keynes said,

“Practical men who believe themselves exempt from any intellectual

influences are usually the slaves of some defunct economist”. The Neo-

classical Approach has become increasingly irrelevant in

understanding of the modern capitalist system.

There are three pervasive economic myths that have been around for a

long time. The first is that markets are fair. Markets do not provide the

right outcomes for the benefit of most of the people. Sometimes, this


163

myth is linked to the idea of ‘invisible hand’ of Adam Smith. But the

situation on which he predicted, it was that there were a large number

of buyers and sellers, none of whom could influence the price and that

those actors were behaving in a moral way. Now neither of these

assumptions applies. We see huge concentration of power in many

markets. Look at the purchasing side. Do people have equal purchasing

rights? The second of the three myths is that prices tell the truth. But,

prices do not tell the social or the environmental truths. Third myth is

that more income equals more happiness. For many people in the world

it does not. It probably holds true if you are poor.

Much of Economics is morally insolvent, because so many people

cling to these old myths. Clearly it suits the most powerful people to

keep perpetuating these myths. The time has come for Economics to

change direction and to find a path which does not deviate from true

human values. The obvious unrealistic nature of Neo-classical

Economics has begun to attract many calls for change. Many

Presidential addresses of the American Economic Association, in

recent times have paid attention on this issue. One of the strongest

agitations regarding this issue has come from university students.

In France, in June 2000, a group of Economics students from some of

the most prestigious universities including Sorbonne published a

petition on the web protesting the lack of realism in Economics

teaching, uncontrolled use and treatment of Mathematics as ‘an end in

itself,’ the repressive domination of Neoclassical Theory and


164

Approaches in university Economics curriculum. They further argued

that the drive to make Economics more like Physics, was flawed and,

that it should be pulled back in line with its more social aspects. They

called Economics they were being taught an ‘autistic science’, lost in

imaginary worlds.

In June 2001, a group of Ph.D. students at Cambridge University

published their petition “Opening up Economics”. By “Opening up

Economics” they meant becoming mindful of the limitations of the

mainstream approach – that is Neo-classical Approach to Economics.

To understand the real world economic issues, they argued that the

validity of the Neo-classical Approach is disputable.

More recently, in November 2011, about 70 undergraduates walked

out from Professor Mankiw’s EC 10 Economics lecture at Harvard,

claiming that his teachings have driven the inequalities in today’s

society. They were deeply concerned about the overly conservative

bias in this Introductory Economics course.

Economics has not experienced such a pressure to change since the

1930s. Then the complaint was its inability to explain the “Great

Depression” and to effect a recovery. It responded by inventing

macroeconomics. Today, indictment is both more general and more

serious: Economics, as taught in universities neither explains

contemporary reality nor provides a framework for the critical debate

of the issues in democratic societies.


165

Does Economics merely study society or does it play a decisive role in

creating it? If we actually construct reality with our thoughts and

actions, then we all have a quite different moral responsibility for what

we say and do. The increased knowledge of ethics should improve the

policy prescriptions of economists. Economic policies, like any other

actions, can be judged by ethical standards. If economists recommend

a policy on economic grounds, but ignore the ethical side of those

policies, they can endorse policies that promote harmful moral

principles, principles that the economist himself might disagree with,

if he reflects on them.

For example, economists often recommend policies on the grounds of

economic efficiency. As long as the efficient use of resources takes

place within a social system, that respects rights, this is a desirable

aspect of economic activity. But what about advocating the poor rural

women to go to Middle East for jobs. This policy may be efficient in

terms of reducing the rate of unemployment and increasing the inward

foreign remittances. Even if these policies are efficient, are they

desirable?

Another advantage of exploring the relationships between Economics

and Ethics is that it would give Economics a broader and more humane

focus. It would help economists focus on the rich contexts within

which economic choice is made. It would emphasize that Economics

is more than the study of maximizing, it is also the study of social order,

which requires acceptance of moral principles. In sum, incorporating

Ethics into Economics will make us more cognizant of the effects of


166

our policy, will improve our theory, and will give Economics a broader

and more humane focus.

REFERENCES:

Boulding, Kenneth (1969) “Economics as a Moral science”, American

Economic review, 59(1), 1-12

Friedman, Milton (1962) Capitalism and Freedom, Chicago:

University of Chicago Press.

Friedman, Milton and Friedman, Rose (1980) Free to Choose, New

York: NY: Harcourt Brace Jovanovich.

Gutenschwager Gerald (2012) “Is Economics a value Free Science?

World Economics Association (WEA) Conferences 2012.

http//weaethicsconferencefiles.worldpress.com. Retrieved

on 10th August 2014.

Robinson, Joan (1977) Morality and Economics, May 1977, University

of Maine. http://economistsview, typepad.com, Retrieved

on July 3, 2014.

Smith Adam (1976a) [1759] The Theory of Moral Sentiments. Edited

by D.D. Raphael and A.L. Macfie. Oxford: Oxford

University Press.

Smith Adam (1976a) [1759]. An Inquiry into the Nature and Causes of

Wealth of Nations. 2 vols. Edited by R.H. Campbell and

A.S Skinner. Oxford: Oxford University Press.

167

BOOK REVIEW


168


169

*Edited by Shunsuke Managi

The Routledge Handbook of

Environmental Economics in

Asia

Routledge New York, 2015

667 pages, Price:

ISBN: 978-0-415-65645-0 (hbk)

ISBN: 978-1-315-74628-9 (ebk)

BOOK REVIEW

Sri Lanka Journal of

Economic Research

Volume 3 (1)

June 2015: 169-178

Sri Lanka Forum of


SLJER

J. G. Sri Ranjith Dept. of Economics and Statistics

University of Peradeniya, SRI LANKA

Telephone: 077-5393462/ 0812392622

Email: [email protected] ,[email protected] ___________________________________________________________

INTRRODUCTION

Environmental economics is rapidly gaining its recognition as a

paramount field of study with the current phase of competitive

development process. The dynamics of such a development phase will

naturally lead to create environmental problems globally, across

countries and regionally or even across provincially within a same

country. The priorities of societies are made more on income,

production and employment than natural resource management and

environmental impacts of such procedures. In their present forms, for

example, almost all the countries national income accounts measure

growth rather than measuring sustainable development. However, as

countries reach their development goals they gradually draw more



170

attention to maintain the quality of environment. This phenomenon has

been correctly observed by the environment Kuznets Curve (EKC)

seems to be a proven fact regarding rich countries as well as middle

income and developing countries. The first comers and later comers to

the industrialization put more emphasis on the importance on keeping

up the developing process with environmental sustainability. Being

one of the fast growing regions countries in Asia except for a few

newly industrialized countries are currently in the process of gaining

their development momentum under the rapid development currently

in progress. Thus it is important to keep on tract with possible adverse

impact and vulnerable effects on environment and consequences in

these countries. Nevertheless the dearth of research, study materials

and textbooks has made some limitations to have a broader and updated

understanding about the subject as a discipline for the students as well

as for the other actors in the field. Hence, the “Routledge Handbook of

Environmental Economics in Asia” is undoubtedly a valuable and

commendable product collection of chapters that covers many of the

vulnerable areas that directly or indirectly link to environment quality

and resource economics under the current phase of development in

Asia-Pacific region. Therefore, it is worth writing a book review on

this publication to provide a better understanding and comments about

the book to the readership and to make the readership even wider. Thus

the main intention of this book review is to uncover some of the

important aspects of using it as a handbook on environmental

economics for teaching, learning and awareness of the subject for

conducting research and policy making.


171

Some attributes of the collection of the chapters and intentions of

the authors

The chapters of the handbook organized in such way as to understand

the evolution/development of the discipline from more traditional

topics of economic growth and environment to tax, emissions trading

and energy utilization to more growing topics including biodiversity,

coastal management and representative country applications. Hence

the sequence of chapters is appropriately arranged and thereby the

reader or the user of the book can either start with the basic concepts

of environmental economics and gradually moving on to contemporary

issues or any specific chapter that reader may think is relevant or

important for a purpose.

The authors are the most recognized well known academics,

researchers and institutional position holders from various parts of the

world are the renowned scholars in this field of study. Collecting book

chapters from such an internationally outstanding scholars in this field

and coordinating them towards getting this final result itself is a great

success and a challenging work. The main intention of writing this

handbook is to provide a prestige reference work to current scholarship

in the expanding discipline of environmental economics particularly

applied in Asia-Pacific region. Achievement of this attribute is evident

with focusing all chapters directly or indirectly to the contemporary or

emerging environmental problems such as coastal development,


172

biodiversity offsets and issues relating to environmental and resource

policy in Asia pacific countries.

As mentioned by the editor of the book this is an intradisciplinary

collection of papers where environmental economics focuses with

wider view to incorporate diverse aspects economy and environmental

problems. This intention of authors reflects from the contents of each

and every chapter as they all have used economic theory, empirical

observations or data collection and analysis to consider the possibility

for actual practice in policy. Furthermore, the lack of literature and

attention to the economic and environment importance with

particularly focus on Asia pacific region identified. Therefore, out of

the 31 chapter more than half is dealt particularly on the countries in

Asia pacific region. The intention of the authors in this aspect seems to

be successful as almost all the chapters have made effort to use

innovative/improved techniques to make those assessments to have

consistency and show directions of the current practices.

Accomplishment of the Purpose of Writing the Book

As shown in the Table of Contents, the volume of 31 chapters is

organized in the order of more traditional topics that provides general

understanding about the subject to diverse and specific areas and

debates with applied techniques for Asia-Pacific. This collection is

therefore appropriately arranged even the beginners to the discipline

will find interesting to follow a course of study in environmental


173

economics since these chapters reflect starting with more general

topics to the progress of increasing diversity in the contemporary

research based on improved techniques. Also each chapter provides a

balanced overview of current knowledge, identifying issues and

discussing relevant debates.

The first five chapters in this handbook clearly shows a general

overview on the relationship between environment and growth which

is relevant not only to Asia pacific but also to any country that making

its effort to achieve development goals. This part starts with exploring

the possible outcomes of climate change and global economy under

three CO2 mitigation scenarios using two analytical methods. It makes

highly relevant for developing countries to draw attention to

environmental quality while development process continuing.

Although there is missing values to the data the third chapter proves

the presence of inverted U shape for Asian countries. This is important

for this region since rapid population growth in china, India and south

Asian countries will become major emitters of CO2 and SO2 in Asia.

Therefore environmental friendly technology is required to mitigate

the impact. It emphasizes in the 4th chapter proving the need for

formulating public policy for making the development process

efficiency in energy intensity and energy efficiency. It poses a question

on to reconsider GHG emission in consumption and production in

chapter 5. This part of the book provides the essentially necessary

background for the students, academics and researchers to prepare and


174

move forward into more specific characteristics and problematic areas

in the discipline.

The second part of the theme identified from this book review goes

from chapter 6 to 20 which spans environmental policy issues and

mitigation measures. For instance the chapter 8 raises the point

correctly with the theory of emissions trading and taxations which

shows commendable effort to formulate good mechanism to policy

making for any country but for the Asian countries market power is

problematic due to monopoly, welfare and distortions. Thus

applicability is problematic due to market inefficiency and uncertainty

with ownership and payment problems. Furthermore, the emerging

situation becomes more challenging since supply and demand trends

of nonrenewable energy in Asian countries with their industrialization,

consumption patterns, urbanization, and wealth and income levels

living standards improving while population keeps expanding.

However, the book chapters 15 and 16 carefully examine the

possibility for addressing these dimensions. The necessary technical

knowledge and methods are presented in terms of valuation of

environmental benefits in Asia using different valuation techniques

that are important for research. Revealed preference method, travel

cost method, hedonic pricing method, averting behavior method,

benefit transfer method, contingent valuation method. In fact these

methods are useful for conducting academic research.

This part of the subject could raise some conflicting debates due to the

diversity in terms of socioeconomic, cultural and political concerns


175

especially in Asian region. The book has drawn attention exploring

the nature of intractable issues in agriculture and tourism industry

where natural resource use and production intensity pose questions on

possibility to apply the valuation techniques and priorities for the

conservation of natural and bio diversity. The problematic situation

and the mitigation measures in this regard is appropriately given in

chapter 19 and 20 referring to faster rate of development in coastal

areas so that damage done to coral reefs, and wildlife. A large

proportion of population is living closer to coastal areas due to many

reasons. Therefore coastal development and protecting ecosystem is

important. Thus authors suggest that tourism can be used as a

conservation tool which is a commendable strategy to mitigate the

adverse impacts. The recommendations and the direct actions implies

by these chapters may not be feasible and politically desirable in the

short run. Nevertheless educating the people improve awareness and

introducing new technological methods of production can help to

mitigate adverse impacts in the long run.

The rest of the chapters are dealing with specific policy issues and

contextual environmental problems plus mitigation measures. The

contents of the chapters explore agreeably the diversity and the rigor

of the environmental problems that depending on the status of the

development in each country and socioeconomic and cultural factors.

In chapter 24 refers inequality and the environment where mostly the

developing countries in Asian experience economic growth with

growing inequality that leading to resulted in greater environmental


176

degradation. Whereas Japan, Korea and China the situation somewhat

different regarding the determination of standards of domestic

environmental policy. These examples and cases provide a broader

understanding about the nature of this discipline to the reader which is

a highly appreciable aspect of this handbook.

Some Comments on the Features of the Book

The first part of the book deals with environmental and resource

economics in general applicable not only in Asia-Pacific regions but

also in other countries regardless of their development status. The book

itself shows that most of environmental damages are created not by

developing countries but by the industrialized countries. This includes

CO2 and SO2 emissions, GHG, E-waste which are tremendously high

in extend compare with the countries in Asian region. However, these

aspects are lightly referred in this book and less attention is paid to

formulate feasible solutions and to explore potentials for some

compromise solution among countries such as Japan, China or any

other industrialized country. However the geopolitical perspectives

relating to environmental problems are not adequately focused in this

book. But such moves may play a crucial role to keep up with

environmental protocols among the countries in Asia and globally for

negotiations.

Also some various methods used in countries to estimate carbon tax,

fuel tax and so on but a little or no attention is drawn on how this tax


177

revenue should be spent on improving the quality of environment.

Therefore, inclusion of some comments and chapters in relation to

implementation of suggested policies would have contributed

immensely to achieve the purpose of this handbook. In most of the

cases the biggest problem for protecting and conserving the

environment is relating the absence of good institutions and lack of

political will. Another practical problem to look at is the economic

hardships in particular resulted when addressing the environment

problems in developing countries. Thus many limitations to these

recommendations arise unless developing countries are adequately

compensated otherwise their growth potential would be hindered and

will further accentuate the growing divergence between developed and

developing world. These practical aspects are lightly touched in this

book and probably the reader will have to read alternative sources to

understand these complicated nature of the subject. Regarding the

technical aspects, the beginners of the subject may find challenging

understanding the book if the reader does not equip with quantitative

methods adequately. Therefore, laymen and beginners need to undergo

some prerequisites to follow a course using this as a handbook.

CONCLUSION

This handbook provides a good collection of papers which provides

current state of knowledge, debates and relevant literature. Also this

collection will undoubtedly add and enrich the literature available


178

particularly regarding the environmental economics and its concerns in

Asia-Pacific region and filling the gap of inadequate material for

students and researchers. Most of the education institutions will find

this book as a textbook for expanding the courses not only on this

branch in social sciences but also agriculture, engineering and other

pure sciences regarding the relationship between development and

environment. The stakeholders in this filed will find the usefulness of

the valuation and analytical techniques to have a better understanding

of the subject. The readership will mostly consist of not only graduate

and postgraduate students, academics and researchers but also the

practitioners in the ground will also find this to be of a valuable

handbook for various purposes such as studying, conducting research

and policy making.

SLJER

The International Journal Published

by the Sri Lanka Forum of University

Economists

The Sri Lanka Journal of

Economic Research

Editorial Office-2015

The Department of Economics,

University of Ruhuna, Matara,

Sri Lanka.

Telephone: + 94 041 22 22681/

Ext.3202

Web: www.slfue.org


_____________________________________________________________

GUIDELINES FOR CONTRIBUTORS

THE SLJER

Sri Lanka Journal of Economic Research (SLJER) is a refereed international

journal, published by the Sri Lanka Forum of University Economists

(SLFUE). It is published thrice yearly, based on the papers presented at the Sri

Lanka Economic Research Conference, the annual international research

symposium of the SLFUE. In addition to research articles, the Journal accepts

manuscripts of short essays on economic and development policy perspectives

and also book reviews.

The authors assume the responsibility of the accuracy of contents in their

respective articles, and neither the SLFUE, nor the Editorial Board of the

SLJER will bear any responsibility in this regard.

MANUSCRIPT SUBMISSION

Articles are accepted in Sinhala, Tamil and English languages. All submissions

should be made electronically as Microsoft Word documents, to the address of

the Chief Editor of the SLJER: [email protected]

Manuscripts submitted will be subject to rigorous and independent double

blind expert review process administered by the Journal’s Editorial Board, and

the authors will be contacted after the review process is complete. The process

generally takes from one to two months. The Editorial Board’s decision

regarding publication of an article will be final.


SUBMISSION GUIDELINES

Title Page

A separate title page should be provided allowing for blind review. All

identifying information should be featured only on the title page. Ensure that

no information identifying the authors, their institutions, or the institutions

cited in the original research are presented in the abstract, text, tables, or

figures.

List names and affiliations of each author. Include full mailing addresses,

phone numbers, and e-mail contact information of every author. Clearly

identify the corresponding, or lead, author. This person will be responsible for

ushering the manuscript through the review and publication process. The lead

author is responsible for keeping contact information updated.

Abstract

An Abstract not exceeding 100 words in English should be included in the

manuscript. It should review the objectives, significance, results, and main

conclusions of the paper. References should not be cited in the Abstract.

Key Words

Authors should submit a list of six unique and informative key words relating

to the manuscript. Undefined abbreviations or acronyms should be avoided in

the keywords. Editorial Board reserves the right to replace key words if

necessary.

Text

Submit all manuscript files as Word .doc or .docx files.

Manuscripts may be typed in FM-Abhaya (Sinhala), Baamini (Tamil) or Times

New Roman (English) fonts. For English language submissions, use British

spelling, with ‘s’ instead of ‘z’ in relevant words, and not American spelling.

Use 12 pt. font size for the entire document, including tables, figures, headings,

captions, and table notes (page footnotes, generated in Word, will be smaller).

Double space the manuscript.

Keep 1-inch margins all round.

Place page number at the bottom centre.

Do not use section or page breaks in the running text.

Use footnotes wherever necessary, but never to solely cite a reference.

Equations

The author is responsible for ensuring that equations are presented properly.

Use Math Type, and not the Word equation function.

Tables

All tables and figures should be professionally typeset based on a standardised

style.

Number all tables (Arabic) as they appear in the text and ensure that all tables

are mentioned in the text

Use Word table generator; do not apply styles, shading, or underlining.

Do not merge cells for spanning headers.

Single space all rows; do not allow lead or following spaces in the paragraph

option.

Do not centre columns or rows.

Use superscript a, b, c, and so forth sparingly for table notes.

Figures

Number all figures (Arabic) as they appear in the text.

Do not use colour in graphics; all pages are printed in black and white. To

distinguish bars or lines, use patterns or very distinct shading.

The author is responsible for securing permission to use graphics or data from

other sources.

If re-creating a graphic from another source, secure permission to adapt the

figure.

Do not send a scan or photocopy of a graphic.

Appendices

Sometimes appendices help readers to further study the issue presented.

Submit appendices separately after attaching tables and figures. If more than

one appendix is featured, use capital letters to refer to them in the text:

Appendix A, Appendix B, etc. Follow guidelines for tables or figures, as

appropriate, when formatting appendices.

References

The detailed style of referencing is as follows:

Journal article

David, J.K. (2010). Reference style guidelines. Journal of Guidelines, 4, 2-7.

Book

David, J.K. (2010). Reference style guidelines. 2nd ed. London: Longman.

Article in an edited book

David, J.K. (2010). Do not capitalise prepositions. In R. Brown (Ed.),

Reference style guidelines. (pp. 55-62). London: Longman.

Edited book

David, J.K. (Ed.). (2010). Reference style guidelines. London: Longman.

Dissertation (unpublished)

David, J.K. (2010). Reference style guidelines. Unpublished doctoral

dissertation, London School of Economics, London.

Article presented at a symposium or annual meeting

David, J.K. (2010, January). A citation for every reference, and a reference for

every citation. Article presented at the annual meeting of the Reference

Guidelines Association, St. Tomas, London.

Online reference

David, J.K. (2010, January). Quotes of 45 or more words will be block quotes.

Reference style guidelines. Retrieved Date, Month, Year, from

http://www.londonpub.com

http://www.londonpub.com/

Documents

SRI LANKA JOURNAL OF ECONOMIC RESEARCHslfue.org/images/SLFUE_downloads/SLJER_Issues/2015June/... · 2016-09-20 · Sri Lanka Journal of Economic Research The International Journal